{ "scenes": { "apple/575_84738_167639": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007127789630827533, "visible": 0.006932789700971948, "excluded": 0.027494334146081226, "within_5pct": 0.9868919456591384, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005847881458527684, "visible": 0.0054612204786167315, "excluded": 0.04791848970340103, "within_5pct": 0.9793562687770205, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6820805696770549, "peak_allocated_gib": 7.612269878387451, "reference_discrepancy": 0.008081953506916761, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008236601544581322, "visible": 0.0069642171890733524, "excluded": 0.11812150913023006, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.014828637325461696, "visible": 0.01068640886514892, "excluded": 0.1811644087320042, "within_5pct": 0.9920223836313304, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 2.166013372130692, "peak_allocated_gib": 9.980838298797607, "reference_discrepancy": 0.014186021871864796, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005541459642473886, "visible": 0.005414421502995466, "excluded": 0.07000320058892096, "within_5pct": 1.0, "coverage": 0.9962138572949665, "pair_records": 18 }, "background": { "discrepancy": 0.006430809789793143, "visible": 0.0063332652852083286, "excluded": 0.0326146568440284, "within_5pct": 0.9969297684031261, "coverage": 0.985146280060145, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 2.924076029099524, "peak_allocated_gib": 8.191169738769531, "reference_discrepancy": 0.004864601884037256, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005414224575363326, "visible": 0.004755610093362771, "excluded": 0.05489828173415483, "within_5pct": 0.9990225161248364, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004080224864064123, "visible": 0.0036998379416722017, "excluded": 0.05080183715360812, "within_5pct": 0.9905674143732106, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.0249873204156756, "peak_allocated_gib": 3.3859000205993652, "reference_discrepancy": 0.00830497033894062, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.015401865325039107, "visible": 0.013148484820573004, "excluded": 0.15299482452584295, "within_5pct": 0.9886234795558271, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.017838441278479403, "visible": 0.013355357765118054, "excluded": 0.02827009305728885, "within_5pct": 0.9911936271340216, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.5136601300910115, "peak_allocated_gib": 6.35312032699585, "reference_discrepancy": 0.0163800953887403, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0063935988712119, "visible": 0.005881730735514677, "excluded": 0.08213967966436667, "within_5pct": 0.9942926665918117, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003874830746690938, "visible": 0.0031835449283763263, "excluded": 0.02468152848966903, "within_5pct": 0.9935699336181845, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5136601300910115, "peak_allocated_gib": 6.35312032699585, "reference_discrepancy": 0.0163800953887403, "rotation_degrees": 1.3099079132080078, "center_rmse": 0.026336530223488808, "focal_difference": 0.03316694498062134 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.0064578463065819176, "visible": 0.0059006367354601935, "excluded": 0.04927043176367257, "within_5pct": 0.9874560305480318, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008228876006779826, "visible": 0.007465696948204108, "excluded": 0.043696204532012964, "within_5pct": 0.9906586940127029, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.4969583004713058, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 2.817288875579834, "center_rmse": 0.05583242326974869, "focal_difference": 0.03337442874908447 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.004718477375970845, "visible": 0.0037753644633875, "excluded": 0.1080181385315386, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.002374143130116182, "visible": 0.002338636096627801, "excluded": 0.02903649842969716, "within_5pct": 0.9964981530736863, "coverage": 0.9957016436900226, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 3.391977434977889, "peak_allocated_gib": 8.074109554290771, "reference_discrepancy": 0.011302076280117035, "rotation_degrees": 0.6489347219467163, "center_rmse": 0.02597576379776001, "focal_difference": 0.008946478366851807 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0008744419928389958, "pair_records": 18 }, "background": { "discrepancy": 0.0009584936824926873, "pair_records": 18 } }, "reference_discrepancy": 0.0068268682807683945, "inference_seconds": 24.593626259826124, "peak_allocated_gib": 8.191169738769531, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0008548099116510055, "pair_records": 18 }, "background": { "discrepancy": 0.0008846163819278886, "pair_records": 18 } }, "reference_discrepancy": 0.006958874873816967, "inference_seconds": 23.36875396501273, "peak_allocated_gib": 8.074108123779297, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0008996831178785508, "pair_records": 18 }, "background": { "discrepancy": 0.0008186367851534594, "pair_records": 18 } }, "reference_discrepancy": 0.006772811524569988, "inference_seconds": 18.831828504800797, "peak_allocated_gib": 8.074108123779297, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0031601393115626318, "visible": 0.00282634118734982, "excluded": 0.07148864667305468, "within_5pct": 0.9967136617607404, "coverage": 0.4966672863301087, "pair_records": 18 }, "background": { "discrepancy": 0.003309979237248608, "visible": 0.003309979237248608, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0009839030770369137, "pair_records": 2 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "bottle/598_92195_184711": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.027933651308897647, "visible": 0.016283302039424172, "excluded": 0.3658234417035836, "within_5pct": 0.8731518475690085, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006005047321291566, "visible": 0.005659029174782362, "excluded": 0.02169357518058348, "within_5pct": 0.9961835676935309, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.474869254976511, "peak_allocated_gib": 7.613957405090332, "reference_discrepancy": 0.04102535545825958, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.023982785473946833, "visible": 0.015026764776393333, "excluded": 0.33785063321685777, "within_5pct": 0.8632739411721861, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0197484457827519, "visible": 0.015281952843765803, "excluded": 0.23437611821980325, "within_5pct": 0.9821425387024509, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.9063176736235619, "peak_allocated_gib": 9.980838298797607, "reference_discrepancy": 0.0358438715338707, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.022429813320932603, "visible": 0.010309834278167194, "excluded": 0.3454156163454244, "within_5pct": 0.814607236703975, "coverage": 0.9999804144306474, "pair_records": 18 }, "background": { "discrepancy": 0.002277899592584239, "visible": 0.002183511595057929, "excluded": 0.023066072238060356, "within_5pct": 0.9930999073185536, "coverage": 0.9835428413330891, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6267429850995541, "peak_allocated_gib": 8.193103313446045, "reference_discrepancy": 0.013480377849191427, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.05068325587926219, "visible": 0.04218354240427932, "excluded": 0.33049537055771117, "within_5pct": 0.5285236280817383, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0049070376708426865, "visible": 0.004423783435298299, "excluded": 0.05696879698207878, "within_5pct": 0.9803245963839471, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.5799436746165156, "peak_allocated_gib": 3.3859000205993652, "reference_discrepancy": 0.02957549411803484, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.033468893322777624, "visible": 0.016255785370533078, "excluded": 0.3436689772972658, "within_5pct": 0.7637945994181556, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004958846473606091, "visible": 0.004873047619059817, "excluded": 0.017873500788971458, "within_5pct": 0.9931826695085675, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3486733725294471, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02766804862767458, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03149688595417014, "visible": 0.01499072250857689, "excluded": 0.34182725145094084, "within_5pct": 0.7663479446799846, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003319010458968894, "visible": 0.0032821535074727074, "excluded": 0.01742155027197858, "within_5pct": 0.9923645790929752, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3486733725294471, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02766804862767458, "rotation_degrees": 0.6496782302856445, "center_rmse": 0.013213261030614376, "focal_difference": 0.018841266632080078 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.030076517153142766, "visible": 0.01794985591083125, "excluded": 0.3567000149219748, "within_5pct": 0.8647540475429192, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006931015644288189, "visible": 0.006527283151242793, "excluded": 0.043463640074303686, "within_5pct": 0.9947578428264413, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.1746012596413493, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 2.3333115577697754, "center_rmse": 0.038960617035627365, "focal_difference": 0.06902772188186646 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.025499735973769165, "visible": 0.013847621370040548, "excluded": 0.3657991942529511, "within_5pct": 0.8382243368788549, "coverage": 0.9999809683313032, "pair_records": 18 }, "background": { "discrepancy": 0.002724643563543738, "visible": 0.0026765570081573823, "excluded": 0.01843337344058222, "within_5pct": 0.9944657423420348, "coverage": 0.9784868693541061, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5597364343702793, "peak_allocated_gib": 8.075524806976318, "reference_discrepancy": 0.015375570859760046, "rotation_degrees": 0.4337581694126129, "center_rmse": 0.008421418257057667, "focal_difference": 0.006591349840164185 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.016363856350068155, "pair_records": 18 }, "background": { "discrepancy": 0.0009585960535068395, "pair_records": 18 } }, "reference_discrepancy": 0.012892407365143299, "inference_seconds": 24.977528943680227, "peak_allocated_gib": 8.191169738769531, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.018124259717047998, "pair_records": 18 }, "background": { "discrepancy": 0.0009046825481350414, "pair_records": 18 } }, "reference_discrepancy": 0.013163352850824594, "inference_seconds": 21.129583349451423, "peak_allocated_gib": 8.64054536819458, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.01843834143617853, "pair_records": 18 }, "background": { "discrepancy": 0.0008895822271337667, "pair_records": 18 } }, "reference_discrepancy": 0.013112278189510107, "inference_seconds": 18.0225625699386, "peak_allocated_gib": 8.64054536819458, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.006133183659249714, "visible": 0.004698389829827854, "excluded": 0.1213786180557209, "within_5pct": 0.9626145159363426, "coverage": 0.2644782948197893, "pair_records": 18 }, "background": { "discrepancy": 0.001289423711882772, "visible": 0.0012446189136198211, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0025283154990815605, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "chair/579_85558_168937": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.019649056941806995, "visible": 0.011600065803590031, "excluded": 0.3335368574383719, "within_5pct": 0.9358662466874763, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005252902364792266, "visible": 0.00513252754249777, "excluded": 0.02627573704583964, "within_5pct": 0.9902395237731458, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.465522937476635, "peak_allocated_gib": 7.613957405090332, "reference_discrepancy": 0.023150715976953506, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.014656502312777707, "visible": 0.009309935245721887, "excluded": 0.23724319748009087, "within_5pct": 0.94040876250821, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.015462685874151422, "visible": 0.010787835895493175, "excluded": 0.12452951373645682, "within_5pct": 0.9720679618771388, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7190883252769709, "peak_allocated_gib": 9.980838298797607, "reference_discrepancy": 0.03175982553511858, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.011116567779260643, "visible": 0.007627126393056861, "excluded": 0.23827736745262512, "within_5pct": 0.9701987532168515, "coverage": 0.9953251956822398, "pair_records": 18 }, "background": { "discrepancy": 0.004455346999826386, "visible": 0.0044263554257130255, "excluded": 0.033342982779697154, "within_5pct": 0.9840028698229133, "coverage": 0.956032036947775, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6245648739859462, "peak_allocated_gib": 8.193103313446045, "reference_discrepancy": 0.010126000735908747, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.024744721019512275, "visible": 0.013362296900743318, "excluded": 0.2837040194552611, "within_5pct": 0.8541819490501288, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.014189874611030008, "visible": 0.008361159740489617, "excluded": 0.0378638672242647, "within_5pct": 0.9772848834817204, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7613425794988871, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.022233503870666027, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03449033442534927, "visible": 0.018809195238028514, "excluded": 0.36400215321541396, "within_5pct": 0.8234030093948113, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.014303251532617169, "visible": 0.012785644437520345, "excluded": 0.06820931940300842, "within_5pct": 0.9836267671838264, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3487078407779336, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02524741366505623, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.015789025240754307, "visible": 0.008528750595992184, "excluded": 0.3320372724695075, "within_5pct": 0.9382763754988908, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004792261622414928, "visible": 0.004697493981862314, "excluded": 0.02136932435163988, "within_5pct": 0.9866915570347565, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3487078407779336, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02524741366505623, "rotation_degrees": 1.1243081092834473, "center_rmse": 0.03063824772834778, "focal_difference": 0.05809974670410156 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.016206646930689565, "visible": 0.011210809282920672, "excluded": 0.2246009699014285, "within_5pct": 0.8895671513510828, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01070882727824721, "visible": 0.008102837482685976, "excluded": 0.03269937917913407, "within_5pct": 0.9895927947046972, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45321709383279085, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 6.1376190185546875, "center_rmse": 0.05384427309036255, "focal_difference": 0.34594500064849854 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.009458868424840833, "visible": 0.005819816458539861, "excluded": 0.20234946643611418, "within_5pct": 0.9794131129851182, "coverage": 0.9972855050412781, "pair_records": 18 }, "background": { "discrepancy": 0.0023362329203887476, "visible": 0.0019357981902401286, "excluded": 0.03521505550793794, "within_5pct": 0.9891314620634524, "coverage": 0.9583059187909829, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5593226905912161, "peak_allocated_gib": 8.075524806976318, "reference_discrepancy": 0.014483402483165264, "rotation_degrees": 0.7231945991516113, "center_rmse": 0.017348207533359528, "focal_difference": 0.029261231422424316 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.003639341063336376, "pair_records": 18 }, "background": { "discrepancy": 0.000712727367565024, "pair_records": 18 } }, "reference_discrepancy": 0.015907044522464275, "inference_seconds": 21.034175904467702, "peak_allocated_gib": 8.760969161987305, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0034769476416163605, "pair_records": 18 }, "background": { "discrepancy": 0.0007266641291124175, "pair_records": 18 } }, "reference_discrepancy": 0.014045877382159233, "inference_seconds": 22.651545996777713, "peak_allocated_gib": 8.642353534698486, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0034821070465949734, "pair_records": 18 }, "background": { "discrepancy": 0.0007710483137385, "pair_records": 18 } }, "reference_discrepancy": 0.01440997002646327, "inference_seconds": 17.314766586758196, "peak_allocated_gib": 8.642353534698486, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.008747171630752766, "visible": 0.006204406127732824, "excluded": 0.1697368253580631, "within_5pct": 0.9913440072400186, "coverage": 0.559209794817584, "pair_records": 18 }, "background": { "discrepancy": 0.0017117165134823988, "visible": 0.0016480996626842153, "excluded": 0.013388592808248018, "within_5pct": 1.0, "coverage": 0.04997979959302262, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "bicycle/504_72680_141364": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.02669508266754303, "visible": 0.014273784840914502, "excluded": 0.3545726113538519, "within_5pct": 0.8138296042975873, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004211603217529149, "visible": 0.004072112081998621, "excluded": 0.028530309037117455, "within_5pct": 0.9942050137868614, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6197320939972997, "peak_allocated_gib": 7.892227649688721, "reference_discrepancy": 0.021261832676827908, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.018076566095388213, "visible": 0.01012033551217962, "excluded": 0.3336787699439841, "within_5pct": 0.939033815069157, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.03995518875126827, "visible": 0.022029842288400523, "excluded": 0.273847538511993, "within_5pct": 0.7274535623868157, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.9544364586472511, "peak_allocated_gib": 10.892635345458984, "reference_discrepancy": 0.012180602177977562, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01146102006959273, "visible": 0.007072322887167037, "excluded": 0.2520839595194863, "within_5pct": 0.9686552381792104, "coverage": 0.9989765102163226, "pair_records": 18 }, "background": { "discrepancy": 0.0024985334435623746, "visible": 0.0023877673089683863, "excluded": 0.04526800011187794, "within_5pct": 0.9879918058815416, "coverage": 0.9983757419562208, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.845112968236208, "peak_allocated_gib": 8.578879833221436, "reference_discrepancy": 0.007752620615065098, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.024004949584544577, "visible": 0.012959894911552817, "excluded": 0.2837837905948858, "within_5pct": 0.8754044520388613, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007138512269218669, "visible": 0.006650952020026099, "excluded": 0.06655858778746099, "within_5pct": 0.9786978354460678, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.1873408602550626, "peak_allocated_gib": 3.411905288696289, "reference_discrepancy": 0.015653643291443586, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.02271179368776397, "visible": 0.011880126039813001, "excluded": 0.3622953633500106, "within_5pct": 0.8762441934017277, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005485536784140696, "visible": 0.00533440393753386, "excluded": 0.03649868272696077, "within_5pct": 0.9891808658924808, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.4693021774291992, "peak_allocated_gib": 6.491513729095459, "reference_discrepancy": 0.013005685061216354, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.017122609562490926, "visible": 0.009638620067565405, "excluded": 0.2604976357422763, "within_5pct": 0.8937799729821005, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0036494771156536793, "visible": 0.003532609592879662, "excluded": 0.03188965107220705, "within_5pct": 0.9901552822055169, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 384, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4693021774291992, "peak_allocated_gib": 6.491513729095459, "reference_discrepancy": 0.013005685061216354, "rotation_degrees": 0.43946096301078796, "center_rmse": 0.010369722731411457, "focal_difference": 0.022603332996368408 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.0338404339756834, "visible": 0.018824019306333248, "excluded": 0.41875425595788857, "within_5pct": 0.7742298396791638, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005653049751348406, "visible": 0.005503962137266404, "excluded": 0.031126670487422115, "within_5pct": 0.9926934221792405, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 392, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.6079028472304344, "peak_allocated_gib": 7.869582653045654, "reference_discrepancy": null, "rotation_degrees": 0.5645558834075928, "center_rmse": 0.020034339278936386, "focal_difference": 0.043786197900772095 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.01329396729057454, "visible": 0.00791079388864046, "excluded": 0.28567928468631687, "within_5pct": 0.9636184714459007, "coverage": 0.9999390592142634, "pair_records": 18 }, "background": { "discrepancy": 0.002269636551509347, "visible": 0.002151749453305668, "excluded": 0.04693636569172779, "within_5pct": 0.9902135142957675, "coverage": 0.9987548026236549, "pair_records": 18 } }, "views": 10, "resolution": [ 392, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.7551294807344675, "peak_allocated_gib": 8.465087413787842, "reference_discrepancy": 0.01350951474159956, "rotation_degrees": 0.4092611074447632, "center_rmse": 0.00885384064167738, "focal_difference": 0.013726294040679932 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.003457969280234306, "pair_records": 18 }, "background": { "discrepancy": 0.0009499956452260224, "pair_records": 18 } }, "reference_discrepancy": 0.004574083257466555, "inference_seconds": 33.128453854471445, "peak_allocated_gib": 8.579791069030762, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.003406297912068419, "pair_records": 18 }, "background": { "discrepancy": 0.00098528462081127, "pair_records": 18 } }, "reference_discrepancy": 0.0044031895231455564, "inference_seconds": 31.94431541953236, "peak_allocated_gib": 9.032657623291016, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0035113196928464566, "pair_records": 18 }, "background": { "discrepancy": 0.0009771044932056659, "pair_records": 18 } }, "reference_discrepancy": 0.004508387530222535, "inference_seconds": 29.36624951660633, "peak_allocated_gib": 9.032657623291016, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0038277093367991657, "visible": 0.0029499702882024403, "excluded": 0.1528157281288448, "within_5pct": 0.987067840099512, "coverage": 0.5284113885040851, "pair_records": 18 }, "background": { "discrepancy": 0.002790510591007911, "visible": 0.0025625739916549355, "excluded": 0.051076139782457765, "within_5pct": 1.0, "coverage": 0.006145653060265802, "pair_records": 18 } }, "views": 10, "resolution": [ 392, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "hydrant/539_78780_154075": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006767907605834345, "visible": 0.00547766728532921, "excluded": 0.15846785753712672, "within_5pct": 0.9810507113770421, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005778941722745539, "visible": 0.005588718921355688, "excluded": 0.022455800376912343, "within_5pct": 0.9986746280124958, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.283886305987835, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.00699001899920404, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00942952672535996, "visible": 0.0072767577109300935, "excluded": 0.21433183057428973, "within_5pct": 0.9945201777564256, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.020194186733044205, "visible": 0.014732725180601484, "excluded": 0.22393474421815535, "within_5pct": 0.9930243897929849, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.4815977569669485, "peak_allocated_gib": 9.980854034423828, "reference_discrepancy": 0.021962585859000683, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.003893907149768864, "visible": 0.0030786708476724335, "excluded": 0.1312336166140854, "within_5pct": 0.9983310793856022, "coverage": 0.999052742007758, "pair_records": 18 }, "background": { "discrepancy": 0.0021500971185189266, "visible": 0.0021256439226913417, "excluded": 0.007423550171481223, "within_5pct": 0.9970936064374043, "coverage": 0.9988034378717903, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6184340473264456, "peak_allocated_gib": 8.193119049072266, "reference_discrepancy": 0.0029765719082206488, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008902816918716312, "visible": 0.006414943811285696, "excluded": 0.14604351820329498, "within_5pct": 0.9743753432929203, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.010207344041872846, "visible": 0.009664401265694105, "excluded": 0.1063453906801261, "within_5pct": 0.9813823355696203, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7753275660797954, "peak_allocated_gib": 3.3811593055725098, "reference_discrepancy": 0.006642123451456428, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.015948125621666857, "visible": 0.013036256306303935, "excluded": 0.2620673222839506, "within_5pct": 0.988394814602993, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.010478246377956388, "visible": 0.01016060925957768, "excluded": 0.027502766554860066, "within_5pct": 0.9956102772975387, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34908444713801146, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.021738942712545395, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01059229013721441, "visible": 0.007312025291181163, "excluded": 0.1563153727427654, "within_5pct": 0.9936094602479864, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.00443291628292582, "visible": 0.00432439670402399, "excluded": 0.014950155139669519, "within_5pct": 0.9954066464917057, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34908444713801146, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.021738942712545395, "rotation_degrees": 1.06699538230896, "center_rmse": 0.029852407053112984, "focal_difference": 0.09056645631790161 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.014218486251071843, "visible": 0.009880869758006703, "excluded": 0.17202660151241686, "within_5pct": 0.963369602508553, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007119875268930355, "visible": 0.006720233131096946, "excluded": 0.02012537743284555, "within_5pct": 0.9987713707183754, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4691678024828434, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 1.3244857788085938, "center_rmse": 0.030425148084759712, "focal_difference": 0.045262932777404785 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.005433030014499068, "visible": 0.00426733399368711, "excluded": 0.12218660544365365, "within_5pct": 0.9974394225997223, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0033710975075146804, "visible": 0.003346120023150625, "excluded": 0.009371638240507754, "within_5pct": 0.998759378534845, "coverage": 0.9998912814420318, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5600632717832923, "peak_allocated_gib": 8.075540542602539, "reference_discrepancy": 0.016323802061378956, "rotation_degrees": 0.9301620721817017, "center_rmse": 0.016450846567749977, "focal_difference": 0.017985284328460693 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.001123826896891098, "pair_records": 18 }, "background": { "discrepancy": 0.0007359117872562479, "pair_records": 18 } }, "reference_discrepancy": 0.0034126436803489923, "inference_seconds": 23.947795083746314, "peak_allocated_gib": 8.762777328491211, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0010934955752759535, "pair_records": 18 }, "background": { "discrepancy": 0.0007414688854478848, "pair_records": 18 } }, "reference_discrepancy": 0.0033860061084851623, "inference_seconds": 23.3897605529055, "peak_allocated_gib": 8.642719745635986, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0010528063784632526, "pair_records": 18 }, "background": { "discrepancy": 0.000756501219120792, "pair_records": 18 } }, "reference_discrepancy": 0.003426757757551968, "inference_seconds": 21.58891180716455, "peak_allocated_gib": 8.642719745635986, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.001102712538470082, "visible": 0.0010273460512799557, "excluded": 0.09923677762934752, "within_5pct": 0.9987379113152266, "coverage": 0.6242774531199349, "pair_records": 18 }, "background": { "discrepancy": 0.0008842026296247605, "visible": 0.0008662080365010015, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0015600326392716047, "pair_records": 16 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "teddybear/537_78252_152125": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00763580580045987, "visible": 0.006213791394554208, "excluded": 0.1538006919324086, "within_5pct": 0.9423796757189875, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004327562960007634, "visible": 0.004182629148386605, "excluded": 0.006762957545197051, "within_5pct": 0.9991960522148435, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4666867256164551, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.00910397619009018, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006806161942174179, "visible": 0.005196518912536641, "excluded": 0.19738490874421138, "within_5pct": 0.9939103145563993, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.03704932480125513, "visible": 0.017637330606574277, "excluded": 0.3964555916009242, "within_5pct": 0.9706463218363386, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7199793793261051, "peak_allocated_gib": 9.980865955352783, "reference_discrepancy": 0.006104714469984174, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.004410479235909273, "visible": 0.003253956868358999, "excluded": 0.170481870366113, "within_5pct": 0.9989538455717599, "coverage": 0.9924505697289475, "pair_records": 18 }, "background": { "discrepancy": 0.0029992333144698365, "visible": 0.0028886763236842528, "excluded": 0.026987533997063908, "within_5pct": 0.9975197013187634, "coverage": 0.9849114892875654, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6195025769993663, "peak_allocated_gib": 8.19313097000122, "reference_discrepancy": 0.004849609918892384, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008936113309597407, "visible": 0.006772230840481115, "excluded": 0.1778004961777926, "within_5pct": 0.9904755934226568, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005208146951805069, "visible": 0.004829335557643236, "excluded": 0.021845486126702167, "within_5pct": 0.9953960112676654, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.601984973065555, "peak_allocated_gib": 3.3843741416931152, "reference_discrepancy": 0.007339599076658487, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.014080412070408284, "visible": 0.010913645104720062, "excluded": 0.21207627599915096, "within_5pct": 0.9783068990016492, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008470717085687508, "visible": 0.008267719201649943, "excluded": 0.019941349591406365, "within_5pct": 0.9970804654211733, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3645042646676302, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.010407731868326664, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00569838255601981, "visible": 0.004581323814942272, "excluded": 0.17955330483826737, "within_5pct": 0.9790656549760166, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004984028731537082, "visible": 0.004956651307132835, "excluded": 0.01250738630043266, "within_5pct": 0.9972564101152224, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3645042646676302, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.010407731868326664, "rotation_degrees": 0.5631479024887085, "center_rmse": 0.016497259959578514, "focal_difference": 0.0333857536315918 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.011508200531898528, "visible": 0.008020208369554468, "excluded": 0.21732976430079134, "within_5pct": 0.9655775697246114, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006283107929251902, "visible": 0.006235490619061703, "excluded": 0.01189554625021785, "within_5pct": 0.999058795535032, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.0936294253915548, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 1.0417389869689941, "center_rmse": 0.043483804911375046, "focal_difference": 0.016083180904388428 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.006299206510263687, "visible": 0.004498417788644929, "excluded": 0.21784712831595826, "within_5pct": 0.997943057705893, "coverage": 0.999115790255247, "pair_records": 18 }, "background": { "discrepancy": 0.002756108060329482, "visible": 0.0027129690128478803, "excluded": 0.024536464890927076, "within_5pct": 0.9973194488250214, "coverage": 0.9887070268654533, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5607655793428421, "peak_allocated_gib": 8.075552463531494, "reference_discrepancy": 0.016984442248940468, "rotation_degrees": 0.5980695486068726, "center_rmse": 0.011132211424410343, "focal_difference": 0.0032066702842712402 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0014967109096320963, "pair_records": 18 }, "background": { "discrepancy": 0.0007160087343628433, "pair_records": 18 } }, "reference_discrepancy": 0.004288604250177741, "inference_seconds": 25.492923267185688, "peak_allocated_gib": 8.76165771484375, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0015296759942067038, "pair_records": 18 }, "background": { "discrepancy": 0.0007364084651040376, "pair_records": 18 } }, "reference_discrepancy": 0.00436421437188983, "inference_seconds": 26.303514474071562, "peak_allocated_gib": 8.645830631256104, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0015697655785579074, "pair_records": 18 }, "background": { "discrepancy": 0.0007099908659496936, "pair_records": 18 } }, "reference_discrepancy": 0.004322011722251773, "inference_seconds": 22.800600836984813, "peak_allocated_gib": 8.645830631256104, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.002538537841868344, "visible": 0.002161848685413652, "excluded": 0.0993326328077008, "within_5pct": 0.998419312435999, "coverage": 0.6262924439886347, "pair_records": 18 }, "background": { "discrepancy": 0.0022373592007623707, "visible": 0.0022373592007623707, "excluded": 0.034782608695652174, "within_5pct": 1.0, "coverage": 0.001838673010789458, "pair_records": 3 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "backpack/386_45948_91845": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005728454618582394, "visible": 0.005459476010112207, "excluded": 0.021958331028831785, "within_5pct": 0.9988997030245532, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007087255389777053, "visible": 0.006393289040613943, "excluded": 0.029743556785362925, "within_5pct": 0.9941438529355029, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4615904176607728, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.009747567120939493, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008563213201351071, "visible": 0.007317437632725469, "excluded": 0.08976916679172497, "within_5pct": 0.9979475393918564, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.04375117672085907, "visible": 0.03532533283615724, "excluded": 0.36185709441213776, "within_5pct": 0.7050731270140915, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7285839058458805, "peak_allocated_gib": 9.980865955352783, "reference_discrepancy": 0.014494478702545166, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.002347281088077308, "visible": 0.0023180392763302747, "excluded": 0.010847746491286779, "within_5pct": 0.999829232696646, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0022703896763140194, "visible": 0.002150062785878308, "excluded": 0.005133410426508079, "within_5pct": 0.9986465333460126, "coverage": 0.9891120295648317, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.617879101075232, "peak_allocated_gib": 8.19313097000122, "reference_discrepancy": 0.0040941021870821714, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00662110560303625, "visible": 0.006368198215990828, "excluded": 0.023021000993397316, "within_5pct": 0.9973900344670831, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.011006962541543997, "visible": 0.008913059821578556, "excluded": 0.08048375612008737, "within_5pct": 0.9897536769539153, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7573445839807391, "peak_allocated_gib": 3.3817524909973145, "reference_discrepancy": 0.006804752629250288, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007525991989406141, "visible": 0.007516288706723726, "excluded": 0.010287742004858427, "within_5pct": 0.9991467754953922, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008961690622266458, "visible": 0.008709610987936635, "excluded": 0.0664100218140289, "within_5pct": 0.9925169902144839, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.6194138145074248, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02226835396140814, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.004859198813650306, "visible": 0.004804070650529623, "excluded": 0.016615038478248998, "within_5pct": 0.9980570218544464, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005687215431476484, "visible": 0.005237433058232127, "excluded": 0.027963261425884144, "within_5pct": 0.9950062546514022, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.6194138145074248, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.02226835396140814, "rotation_degrees": 2.353221893310547, "center_rmse": 0.03593561425805092, "focal_difference": 0.020390033721923828 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.005980881732160404, "visible": 0.005774994234679571, "excluded": 0.023584972911320216, "within_5pct": 0.9973832671319389, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006337647245404867, "visible": 0.00597145049654509, "excluded": 0.03231694017812465, "within_5pct": 0.9965762432545306, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45190093107521534, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 5.815622329711914, "center_rmse": 0.137587308883667, "focal_difference": 0.04553729295730591 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.002528231452518996, "visible": 0.0024515289430871367, "excluded": 0.011346147165397112, "within_5pct": 0.9993287753311448, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0021181296449535756, "visible": 0.0019431566300242044, "excluded": 0.009669332964331062, "within_5pct": 0.9993407191190369, "coverage": 0.9737442445762448, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.558645986020565, "peak_allocated_gib": 8.075552463531494, "reference_discrepancy": 0.01771312952041626, "rotation_degrees": 1.006190299987793, "center_rmse": 0.022459637373685837, "focal_difference": 0.019181638956069946 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0006681730973138863, "pair_records": 18 }, "background": { "discrepancy": 0.0006695144757580004, "pair_records": 18 } }, "reference_discrepancy": 0.006474449299275875, "inference_seconds": 25.504816045053303, "peak_allocated_gib": 8.762182712554932, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0006573257288662778, "pair_records": 18 }, "background": { "discrepancy": 0.0006414200818243444, "pair_records": 18 } }, "reference_discrepancy": 0.006110438145697117, "inference_seconds": 27.37593425065279, "peak_allocated_gib": 8.64885425567627, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0006722045061118504, "pair_records": 18 }, "background": { "discrepancy": 0.0006621146603431157, "pair_records": 18 } }, "reference_discrepancy": 0.00601030676625669, "inference_seconds": 22.69287585001439, "peak_allocated_gib": 8.64885425567627, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0015897327633826604, "visible": 0.0015702227381556232, "excluded": 0.008856097976529738, "within_5pct": 1.0, "coverage": 0.5015534867242186, "pair_records": 18 }, "background": { "discrepancy": 0.002098540294004607, "visible": 0.0017453957271998523, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.001667900008708945, "pair_records": 10 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "bench/596_91122_180858": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.011326392396249583, "visible": 0.008241926584848215, "excluded": 0.12156509805356083, "within_5pct": 0.9809762360711376, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.0048650615319156855, "visible": 0.0046483858447105095, "excluded": 0.05400504060329814, "within_5pct": 0.993569305365726, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.0583595475181937, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.00971230212599039, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01187572455000848, "visible": 0.00874148498538754, "excluded": 0.23901962699952872, "within_5pct": 0.9910282094744863, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.04950299365427731, "visible": 0.013666985454789376, "excluded": 0.15253709998907822, "within_5pct": 0.9780691484296582, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.711148745380342, "peak_allocated_gib": 9.980865955352783, "reference_discrepancy": 0.026250047609210014, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.003302227011507637, "visible": 0.002964238123734338, "excluded": 0.06241456692484504, "within_5pct": 0.999607586100455, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.0019434820686671382, "visible": 0.0016115098224493672, "excluded": 0.041426799377227354, "within_5pct": 0.9923459928515452, "coverage": 0.9997364970789793, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6186287133023143, "peak_allocated_gib": 8.19313097000122, "reference_discrepancy": 0.003957106499001384, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008893345890333536, "visible": 0.0074625656253588715, "excluded": 0.1821373444501733, "within_5pct": 0.9953901420365618, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.012926851096416857, "visible": 0.007446451125976803, "excluded": 0.18274949704192453, "within_5pct": 0.960987457399711, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.52208421099931, "peak_allocated_gib": 3.3875069618225098, "reference_discrepancy": 0.008326676907017827, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01770662997529865, "visible": 0.0134728199548005, "excluded": 0.14955456864327002, "within_5pct": 0.9957935251830878, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.013944470049575431, "visible": 0.012886147237628511, "excluded": 0.08978593784931813, "within_5pct": 0.9908127750735064, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34930181968957186, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.009281146340072155, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.011816315728442627, "visible": 0.010344585316604257, "excluded": 0.15538165426042516, "within_5pct": 0.9964983202091926, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.00555743183397687, "visible": 0.005003611947986115, "excluded": 0.06897797046494233, "within_5pct": 0.9880721213726009, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34930181968957186, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.009281146340072155, "rotation_degrees": 0.5336868762969971, "center_rmse": 0.012031606398522854, "focal_difference": 0.029238522052764893 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.014347099310146149, "visible": 0.010631649859433809, "excluded": 0.07526205450733753, "within_5pct": 0.9832458847947207, "coverage": 1.0, "pair_records": 15 }, "background": { "discrepancy": 0.007730830289618745, "visible": 0.006487009263066671, "excluded": 0.05708069058326113, "within_5pct": 0.9925909330689394, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4530803533270955, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 2.08160400390625, "center_rmse": 0.04371189326047897, "focal_difference": 0.14142322540283203 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.004878616433616444, "visible": 0.004495888116563407, "excluded": 0.07261735236992492, "within_5pct": 0.9987436680369317, "coverage": 1.0, "pair_records": 14 }, "background": { "discrepancy": 0.003218050275897442, "visible": 0.002708445055892756, "excluded": 0.03910228458078968, "within_5pct": 0.9943339147274224, "coverage": 0.9998689939351257, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5584068289026618, "peak_allocated_gib": 8.075552463531494, "reference_discrepancy": 0.00925210677087307, "rotation_degrees": 0.3520567715167999, "center_rmse": 0.00925295427441597, "focal_difference": 0.018559038639068604 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.001184087152670332, "pair_records": 14 }, "background": { "discrepancy": 0.0010209350617893128, "pair_records": 18 } }, "reference_discrepancy": 0.007291571004316211, "inference_seconds": 26.126074872910976, "peak_allocated_gib": 8.764160633087158, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0012735544473172346, "pair_records": 14 }, "background": { "discrepancy": 0.0010430269799095817, "pair_records": 18 } }, "reference_discrepancy": 0.006817398825660348, "inference_seconds": 23.349745927378535, "peak_allocated_gib": 8.646716117858887, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.001252203850092511, "pair_records": 14 }, "background": { "discrepancy": 0.0009874072405574902, "pair_records": 18 } }, "reference_discrepancy": 0.006722427671775222, "inference_seconds": 24.09587026387453, "peak_allocated_gib": 8.646716117858887, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0014146439036871582, "visible": 0.0013598270288681254, "excluded": 0.023605525441510754, "within_5pct": 1.0, "coverage": 0.5993416112445913, "pair_records": 14 }, "background": { "discrepancy": 0.0017827804962311153, "visible": 0.0017468596129099046, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.010778912456043108, "pair_records": 10 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "bowl/69_5490_12853": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.019038239975214335, "visible": 0.014155792232452169, "excluded": 0.24820747476335153, "within_5pct": 0.9275276840972984, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.038000715527879585, "visible": 0.027761359219235476, "excluded": 0.31637186977680787, "within_5pct": 0.7108308772055432, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6247123414650559, "peak_allocated_gib": 7.892227649688721, "reference_discrepancy": 0.05955776199698448, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00545010411574836, "visible": 0.005196601317724146, "excluded": 0.04962558700555736, "within_5pct": 0.9955887587503562, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.044705052944611606, "visible": 0.02009985783919019, "excluded": 0.25877571598172033, "within_5pct": 0.8544533272831848, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.7329915165901184, "peak_allocated_gib": 10.89266300201416, "reference_discrepancy": 0.006184822879731655, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00315394666715039, "visible": 0.0030049251721953807, "excluded": 0.018989101880793745, "within_5pct": 0.9995362812846256, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0017856021786540814, "visible": 0.001768844329449707, "excluded": 0.007864790203543106, "within_5pct": 0.9973873849530068, "coverage": 0.9993162082930005, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.8412063512951136, "peak_allocated_gib": 8.578907489776611, "reference_discrepancy": 0.003967696335166693, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006576102437705332, "visible": 0.006527287531314976, "excluded": 0.02859322127461618, "within_5pct": 0.9986499222158945, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0063743283009521454, "visible": 0.006254863771621337, "excluded": 0.014245867191236716, "within_5pct": 0.9968670002411608, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.0213676411658525, "peak_allocated_gib": 3.411078453063965, "reference_discrepancy": 0.013095159083604813, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.016346373539238374, "visible": 0.013891263100250812, "excluded": 0.14168208369008412, "within_5pct": 0.9751551285928581, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.018229012802079295, "visible": 0.017001860875669992, "excluded": 0.003302928130655293, "within_5pct": 0.9994206027631201, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 1.139049082994461, "peak_allocated_gib": 6.491513729095459, "reference_discrepancy": 0.030734880827367306, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005381520589055516, "visible": 0.004655234259994466, "excluded": 0.0825927827549667, "within_5pct": 0.978811473696903, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003362604077444694, "visible": 0.003360157077953325, "excluded": 0.001366818621468196, "within_5pct": 0.9997892345008549, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 384 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.139049082994461, "peak_allocated_gib": 6.491513729095459, "reference_discrepancy": 0.030734880827367306, "rotation_degrees": 2.208704948425293, "center_rmse": 0.040070295333862305, "focal_difference": 0.02420628070831299 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.027048581119134642, "visible": 0.015782675465616833, "excluded": 0.21973945321320432, "within_5pct": 0.8536449017602248, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.026830506145714715, "visible": 0.013912481390214941, "excluded": 0.24788076379572896, "within_5pct": 0.9784778865756145, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 392 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.6117929350584745, "peak_allocated_gib": 7.869582653045654, "reference_discrepancy": null, "rotation_degrees": 16.73711395263672, "center_rmse": 0.2095583975315094, "focal_difference": 0.504696249961853 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.004177642739200399, "visible": 0.004106062148611518, "excluded": 0.05088350492601579, "within_5pct": 0.9973360084339258, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.002394773420494382, "visible": 0.0023787948068263965, "excluded": 0.00809508369226558, "within_5pct": 0.997700192077352, "coverage": 0.9972246141906654, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 392 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.7553588636219501, "peak_allocated_gib": 8.465115070343018, "reference_discrepancy": 0.00862726871855557, "rotation_degrees": 0.8662794232368469, "center_rmse": 0.013209793716669083, "focal_difference": 0.004974156618118286 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0009620864360458537, "pair_records": 18 }, "background": { "discrepancy": 0.0005631770232052641, "pair_records": 18 } }, "reference_discrepancy": 0.009080523159354925, "inference_seconds": 28.698302693665028, "peak_allocated_gib": 9.149350643157959, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0009397913513362148, "pair_records": 18 }, "background": { "discrepancy": 0.0005724771914019453, "pair_records": 18 } }, "reference_discrepancy": 0.008699800353497267, "inference_seconds": 30.056043923832476, "peak_allocated_gib": 9.035905838012695, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.00100312281439181, "pair_records": 18 }, "background": { "discrepancy": 0.0005588867158930484, "pair_records": 18 } }, "reference_discrepancy": 0.009203890338540077, "inference_seconds": 26.27892955671996, "peak_allocated_gib": 9.035905838012695, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0016928191869175274, "visible": 0.0016451574414010801, "excluded": 0.009164300626529582, "within_5pct": 0.9984447123443101, "coverage": 0.2977729681758641, "pair_records": 18 }, "background": { "discrepancy": 0.0008694482822754847, "visible": 0.0007579319611773305, "excluded": 0.011111111111111112, "within_5pct": 1.0, "coverage": 0.001437583875231273, "pair_records": 6 } }, "views": 10, "resolution": [ 518, 392 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "toyplane/432_61238_120319": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007273854559842575, "visible": 0.005958268405043687, "excluded": 0.09459704902261876, "within_5pct": 0.9857705392134138, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0030897099335362075, "visible": 0.003070351288476785, "excluded": 0.009950793234617162, "within_5pct": 0.9980438138962086, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4623750727623701, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.009687214158475399, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005845766064612969, "visible": 0.005052426866062706, "excluded": 0.05287296668965945, "within_5pct": 0.997322906236197, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.015043529789242241, "visible": 0.010598739684843855, "excluded": 0.10567950839334407, "within_5pct": 0.9860105580693816, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7185023948550224, "peak_allocated_gib": 9.980881690979004, "reference_discrepancy": 0.007226303918287158, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0024225130093670187, "visible": 0.0022942382677789185, "excluded": 0.040404000218373544, "within_5pct": 0.9984790757030648, "coverage": 0.9998527853020845, "pair_records": 18 }, "background": { "discrepancy": 0.0021798741354198308, "visible": 0.002108585323071328, "excluded": 0.027835257602817694, "within_5pct": 0.9940329476099968, "coverage": 0.9997675035360422, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.4278382752090693, "peak_allocated_gib": 8.193146705627441, "reference_discrepancy": 0.004797583911567926, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00807626372881683, "visible": 0.006974804016751154, "excluded": 0.10127507744416775, "within_5pct": 0.9949819090313345, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008212319093696671, "visible": 0.007599749865102355, "excluded": 0.047542866635410436, "within_5pct": 0.9916526089017743, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7555150743573904, "peak_allocated_gib": 3.3843741416931152, "reference_discrepancy": 0.010985307861119509, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01528620125463487, "visible": 0.01095856873230315, "excluded": 0.12294024484169401, "within_5pct": 0.9937025911734625, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.010120677621996654, "visible": 0.008867681192637245, "excluded": 0.03092909059022974, "within_5pct": 0.9877627157984458, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3486448163166642, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.016643651761114597, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005098726299279365, "visible": 0.004900839683524198, "excluded": 0.047903497120515844, "within_5pct": 0.995999470475305, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.00823692708500181, "visible": 0.008018211900468718, "excluded": 0.03415693218239685, "within_5pct": 0.9920051761827822, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3486448163166642, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.016643651761114597, "rotation_degrees": 1.0065722465515137, "center_rmse": 0.018443860113620758, "focal_difference": 0.037223100662231445 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.007939280171460342, "visible": 0.007004969776896937, "excluded": 0.1074421759072037, "within_5pct": 0.983584518426849, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006156532136200031, "visible": 0.005834574838799825, "excluded": 0.024533707902944568, "within_5pct": 0.9984496371705092, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4525577034801245, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 3.2884726524353027, "center_rmse": 0.034574609249830246, "focal_difference": 0.1797102689743042 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.004209251817381252, "visible": 0.004030769154869201, "excluded": 0.04888081377530993, "within_5pct": 0.9985412396809008, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004188465537103121, "visible": 0.004011856988050334, "excluded": 0.025801268487439667, "within_5pct": 0.9928960989514195, "coverage": 0.9998995305274176, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5585838723927736, "peak_allocated_gib": 8.075568199157715, "reference_discrepancy": 0.00813542352989316, "rotation_degrees": 0.28039300441741943, "center_rmse": 0.008669133298099041, "focal_difference": 0.012216627597808838 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0011222369314423516, "pair_records": 18 }, "background": { "discrepancy": 0.0009739286464952961, "pair_records": 18 } }, "reference_discrepancy": 0.010070452932268381, "inference_seconds": 26.90542967710644, "peak_allocated_gib": 8.769418716430664, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0011133302330487694, "pair_records": 18 }, "background": { "discrepancy": 0.0009903239492013815, "pair_records": 18 } }, "reference_discrepancy": 0.010810436215251684, "inference_seconds": 26.051685725338757, "peak_allocated_gib": 8.651564121246338, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0011175681645854627, "pair_records": 18 }, "background": { "discrepancy": 0.0009966722311286194, "pair_records": 18 } }, "reference_discrepancy": 0.00987891387194395, "inference_seconds": 22.989641718566418, "peak_allocated_gib": 8.651564121246338, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0028161790973716175, "visible": 0.0026403006485606604, "excluded": 0.03643038047563871, "within_5pct": 0.9981660772423544, "coverage": 0.6111498978888054, "pair_records": 18 }, "background": { "discrepancy": 0.0029869831998958754, "visible": 0.0027191881450752, "excluded": 0.007878411910669975, "within_5pct": 1.0, "coverage": 0.004312634992124838, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "motorcycle/582_86621_170494": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.034310586292605055, "visible": 0.01817928589139002, "excluded": 0.3799807923531848, "within_5pct": 0.7838881404601661, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0089772993342911, "visible": 0.008353100684547865, "excluded": 0.0825616651502637, "within_5pct": 0.9862976234330174, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.5442726612091064, "peak_allocated_gib": 7.612269878387451, "reference_discrepancy": 0.029278980568051338, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03155209433231234, "visible": 0.015185942906208765, "excluded": 0.35402561912107877, "within_5pct": 0.8414701209700475, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.03714607053966967, "visible": 0.018626195580642598, "excluded": 0.2774989736388653, "within_5pct": 0.6523248093676438, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 2.1172451712191105, "peak_allocated_gib": 9.980838298797607, "reference_discrepancy": 0.03752797469496727, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0182432466884083, "visible": 0.010385921257091922, "excluded": 0.22141163008802772, "within_5pct": 0.9504231171085689, "coverage": 0.9999510274733243, "pair_records": 18 }, "background": { "discrepancy": 0.006186147495841456, "visible": 0.004748466415973039, "excluded": 0.09630840927380599, "within_5pct": 0.9754190256544746, "coverage": 0.9849232407202126, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 3.0895020458847284, "peak_allocated_gib": 8.191169738769531, "reference_discrepancy": 0.016799391247332096, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.034176924169577, "visible": 0.018055931609951593, "excluded": 0.4261049198558315, "within_5pct": 0.8072492825200375, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.017720999806131815, "visible": 0.01155312816541899, "excluded": 0.17891693236560058, "within_5pct": 0.9377256466925317, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.046495753340423, "peak_allocated_gib": 3.3859000205993652, "reference_discrepancy": 0.024828686378896236, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03350906106309533, "visible": 0.015136928366538922, "excluded": 0.29741989792427104, "within_5pct": 0.8607201253338788, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.013297165787334597, "visible": 0.010980745914486136, "excluded": 0.119679276179848, "within_5pct": 0.9529555854313015, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 1.3135299123823643, "peak_allocated_gib": 6.35312032699585, "reference_discrepancy": 0.031856270506978035, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.026142886554151177, "visible": 0.012468286641402046, "excluded": 0.3186431450001628, "within_5pct": 0.8882212392373597, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008180672052289454, "visible": 0.00720782176283692, "excluded": 0.11085113123801629, "within_5pct": 0.9674271166215519, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.3135299123823643, "peak_allocated_gib": 6.35312032699585, "reference_discrepancy": 0.031856270506978035, "rotation_degrees": 1.936939001083374, "center_rmse": 0.025003308430314064, "focal_difference": 0.0692242980003357 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.03836446624533025, "visible": 0.017074317278066475, "excluded": 0.41035439073759483, "within_5pct": 0.8070740246752326, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.009379988953921418, "visible": 0.008527577745066526, "excluded": 0.07330686511892893, "within_5pct": 0.9782827012162616, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.649817880243063, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 5.223941802978516, "center_rmse": 0.09633812308311462, "focal_difference": 0.3858933448791504 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.01570155175916422, "visible": 0.010089013059113552, "excluded": 0.2378863225400291, "within_5pct": 0.9455949619901833, "coverage": 0.9999489613637524, "pair_records": 18 }, "background": { "discrepancy": 0.006323649751750289, "visible": 0.005212420603573199, "excluded": 0.11037777443693633, "within_5pct": 0.974382279217272, "coverage": 0.9804859749487967, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.610251970589161, "peak_allocated_gib": 8.074109554290771, "reference_discrepancy": 0.030339185148477554, "rotation_degrees": 1.2299771308898926, "center_rmse": 0.0236850343644619, "focal_difference": 0.007886707782745361 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.004600443207653352, "pair_records": 18 }, "background": { "discrepancy": 0.0018872416972575577, "pair_records": 18 } }, "reference_discrepancy": 0.021156156435608864, "inference_seconds": 21.85824563819915, "peak_allocated_gib": 8.76793098449707, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.005049690542228617, "pair_records": 18 }, "background": { "discrepancy": 0.0017680256599793753, "pair_records": 18 } }, "reference_discrepancy": 0.022114346735179424, "inference_seconds": 22.38242146372795, "peak_allocated_gib": 8.64835786819458, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.005178465403423911, "pair_records": 18 }, "background": { "discrepancy": 0.0017906824225996335, "pair_records": 18 } }, "reference_discrepancy": 0.022218523547053337, "inference_seconds": 18.703205971978605, "peak_allocated_gib": 8.64835786819458, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.013079889620347524, "visible": 0.006507377410734752, "excluded": 0.31817916850548655, "within_5pct": 0.9797606695794578, "coverage": 0.3580713440566255, "pair_records": 18 }, "background": { "discrepancy": 0.005577321558771186, "visible": 0.0036241707026593204, "excluded": 0.11290322580645161, "within_5pct": 1.0, "coverage": 0.0035978346752978466, "pair_records": 9 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "car/194_20939_43630": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03805076846651889, "visible": 0.012403109423474053, "excluded": 0.3087034780911464, "within_5pct": 0.9249012847278286, "coverage": 0.9982284185696957, "pair_records": 18 }, "background": { "discrepancy": 0.01262887691138229, "visible": 0.01175228220043582, "excluded": 0.12656090444015686, "within_5pct": 0.9647375997428446, "coverage": 0.8474186801938006, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4741717092692852, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.015671168453991413, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.011660493065994783, "visible": 0.00682618938600062, "excluded": 0.23282043340016673, "within_5pct": 0.9595186102451116, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.03216723621124164, "visible": 0.0172665022449175, "excluded": 0.2393287343054819, "within_5pct": 0.9113386613386614, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.997022190131247, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.009458203800022602, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0145214731827991, "visible": 0.006644460430244625, "excluded": 0.2983170253674788, "within_5pct": 0.9530965529797306, "coverage": 0.9976428303130787, "pair_records": 18 }, "background": { "discrepancy": 0.006582688894600213, "visible": 0.005574564602531118, "excluded": 0.08326095808968449, "within_5pct": 0.9874652448634882, "coverage": 0.7785986044484691, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6208205474540591, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.008416688069701195, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.06163336843439148, "visible": 0.021879921770557035, "excluded": 0.4988673682065249, "within_5pct": 0.6896840737465555, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.026017887637394776, "visible": 0.018690237713243664, "excluded": 0.2546364119870667, "within_5pct": 0.8723181333782041, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.8733647409826517, "peak_allocated_gib": 3.3859000205993652, "reference_discrepancy": 0.022889153100550175, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.05451592681454011, "visible": 0.019098763092741198, "excluded": 0.6598526894076964, "within_5pct": 0.9081190112009244, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.021574530258575433, "visible": 0.014626577042490731, "excluded": 0.21357929136566906, "within_5pct": 0.932514227073421, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 1.1425434881821275, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.03967371769249439, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.022157621055187, "visible": 0.014967475040021556, "excluded": 0.41757783758451333, "within_5pct": 0.9584813146084881, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.012618601480840316, "visible": 0.00908096418584248, "excluded": 0.1638103377692754, "within_5pct": 0.9543774919199639, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.1425434881821275, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.03967371769249439, "rotation_degrees": 0.5845028162002563, "center_rmse": 0.02101716957986355, "focal_difference": 0.04365748167037964 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.034877575226604424, "visible": 0.012016799670361818, "excluded": 0.37763048017476286, "within_5pct": 0.9109358122285363, "coverage": 0.9987113039289524, "pair_records": 18 }, "background": { "discrepancy": 0.015180903154251718, "visible": 0.009807479294403931, "excluded": 0.1667341860061346, "within_5pct": 0.9310149263348537, "coverage": 0.8625983437060181, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4729841537773609, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 1.4411215782165527, "center_rmse": 0.01349404826760292, "focal_difference": 0.0744246244430542 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.017256175399025894, "visible": 0.009294778636394468, "excluded": 0.3247956760172222, "within_5pct": 0.9512450832752948, "coverage": 0.9972294055784042, "pair_records": 18 }, "background": { "discrepancy": 0.007184770073604798, "visible": 0.006290642754114789, "excluded": 0.07533070585782005, "within_5pct": 0.9711480908152734, "coverage": 0.7713007504475922, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5597020220011473, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.025801152922213078, "rotation_degrees": 0.5178661346435547, "center_rmse": 0.01882052607834339, "focal_difference": 0.022387564182281494 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.005869264359726839, "pair_records": 18 }, "background": { "discrepancy": 0.0035964558751551716, "pair_records": 18 } }, "reference_discrepancy": 0.006908384384587407, "inference_seconds": 19.769183043390512, "peak_allocated_gib": 8.768982887268066, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.005383438248997132, "pair_records": 18 }, "background": { "discrepancy": 0.0035106972676198153, "pair_records": 18 } }, "reference_discrepancy": 0.006542573217302561, "inference_seconds": 20.621640348806977, "peak_allocated_gib": 8.652052402496338, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.005049929254719164, "pair_records": 18 }, "background": { "discrepancy": 0.0034891484483019946, "pair_records": 18 } }, "reference_discrepancy": 0.00677984394133091, "inference_seconds": 18.009897467680275, "peak_allocated_gib": 8.652052402496338, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.00855482343184138, "visible": 0.004070370497615473, "excluded": 0.23418461047628528, "within_5pct": 0.9951390069314597, "coverage": 0.20880778984055537, "pair_records": 18 }, "background": { "discrepancy": 0.005593749950024644, "visible": 0.00508797601225791, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0022268520231193134, "pair_records": 3 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "plant/571_83189_164325": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.04709277028141992, "visible": 0.021478694673304224, "excluded": 0.42388689308575883, "within_5pct": 0.6907280997286842, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.00270631479481751, "visible": 0.0026826925569145613, "excluded": 0.013660008730164378, "within_5pct": 0.9979825475502985, "coverage": 0.9988127760614803, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4804415125399828, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.044333815574645996, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.04379976781742079, "visible": 0.01889876916430449, "excluded": 0.44303947653518333, "within_5pct": 0.7331183672421209, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.019651245095297733, "visible": 0.013990490294805133, "excluded": 0.2319144833622563, "within_5pct": 0.9252245058972234, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6913112979382277, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.03140265308320522, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03344429337894498, "visible": 0.014076998859914027, "excluded": 0.42835390603464185, "within_5pct": 0.840152232228549, "coverage": 0.9993866636419244, "pair_records": 18 }, "background": { "discrepancy": 0.0029877068296557683, "visible": 0.002960229825991257, "excluded": 0.029081870528481706, "within_5pct": 0.9804985042220076, "coverage": 0.9560181032247512, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6238616276532412, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.01971473079174757, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.04078765569704518, "visible": 0.018067980327471516, "excluded": 0.4065811270683309, "within_5pct": 0.7294453972777291, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.002492834535014067, "visible": 0.0023786568713808154, "excluded": 0.04253498185093878, "within_5pct": 0.9762997056891737, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 1.5115881441161036, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.03640383668243885, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.04784419383313615, "visible": 0.019499235537225275, "excluded": 0.4357679714005507, "within_5pct": 0.6982198973317979, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006422823686419962, "visible": 0.006310781832383435, "excluded": 0.04813452510251068, "within_5pct": 0.9870585665030109, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34855112060904503, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.037632543593645096, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03944612705456453, "visible": 0.01787055684363214, "excluded": 0.4095926623969449, "within_5pct": 0.6928700104072818, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003257388266369681, "visible": 0.0030576505521340116, "excluded": 0.040334283254629844, "within_5pct": 0.9840802266694157, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34855112060904503, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.037632543593645096, "rotation_degrees": 0.7313711643218994, "center_rmse": 0.013839688152074814, "focal_difference": 0.04815250635147095 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.04125439263176521, "visible": 0.018742354386500945, "excluded": 0.4381667949140181, "within_5pct": 0.7341400370233282, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004399142210191153, "visible": 0.004358712847545892, "excluded": 0.020684197723826334, "within_5pct": 0.9935422769043247, "coverage": 0.9990150503601867, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 1.1423005871474743, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 1.5777817964553833, "center_rmse": 0.014636101201176643, "focal_difference": 0.11819928884506226 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.03630825920969845, "visible": 0.013401548303499534, "excluded": 0.43032319291578214, "within_5pct": 0.8421546737817965, "coverage": 0.9997347282924827, "pair_records": 18 }, "background": { "discrepancy": 0.002358694645164087, "visible": 0.002309740538114057, "excluded": 0.02553350030228163, "within_5pct": 0.9829849263234145, "coverage": 0.9518634681816476, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5589811028912663, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.020460885018110275, "rotation_degrees": 0.4518343508243561, "center_rmse": 0.00723958108574152, "focal_difference": 0.002922147512435913 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.02202651131846725, "pair_records": 18 }, "background": { "discrepancy": 0.0011300380775979332, "pair_records": 18 } }, "reference_discrepancy": 0.01634243782609701, "inference_seconds": 20.871275468729436, "peak_allocated_gib": 8.190349102020264, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.02285580343159263, "pair_records": 18 }, "background": { "discrepancy": 0.0010977452503890181, "pair_records": 18 } }, "reference_discrepancy": 0.016192362643778324, "inference_seconds": 23.24608540534973, "peak_allocated_gib": 8.648127555847168, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.021456336251758884, "pair_records": 18 }, "background": { "discrepancy": 0.001067964003921893, "pair_records": 18 } }, "reference_discrepancy": 0.015681021381169558, "inference_seconds": 17.87025231309235, "peak_allocated_gib": 8.648127555847168, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.008216547911109755, "visible": 0.004072777410754081, "excluded": 0.28917441682732464, "within_5pct": 0.9186369223111781, "coverage": 0.6197661291831749, "pair_records": 18 }, "background": { "discrepancy": 0.002777279800794753, "visible": 0.002508380959736766, "excluded": 0.03425854037267081, "within_5pct": 1.0, "coverage": 0.006610187278789962, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "umbrella/618_100194_200292": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.016581407752292533, "visible": 0.011221050693714896, "excluded": 0.21907693571632308, "within_5pct": 0.9563508903360757, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.010794815903388402, "visible": 0.009152063098533864, "excluded": 0.11184863523573202, "within_5pct": 0.9740096313082891, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4746630946174264, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.020168215036392212, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01619232917525821, "visible": 0.01220446621046278, "excluded": 0.21301519477374292, "within_5pct": 0.9868450876561574, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.10654013199035485, "visible": 0.0881792134152525, "excluded": 0.4223900199649495, "within_5pct": 0.34707971586424624, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6766847837716341, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.017440328374505043, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007101722979153713, "visible": 0.005977486122682906, "excluded": 0.0752401534335142, "within_5pct": 0.9984296539029018, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004584352160872255, "visible": 0.003779428604936942, "excluded": 0.07228675880326604, "within_5pct": 0.9788855333025845, "coverage": 0.9925537749243807, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6332281231880188, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.008684123400598764, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.012832798614017663, "visible": 0.009672230432081211, "excluded": 0.19978971490335512, "within_5pct": 0.9705759167463266, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.019527619003552245, "visible": 0.013147054116215228, "excluded": 0.19352040366649037, "within_5pct": 0.9094631815380481, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.8147478923201561, "peak_allocated_gib": 3.381706714630127, "reference_discrepancy": 0.022179565392434597, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01338767142227527, "visible": 0.010578230891525467, "excluded": 0.09113436920040083, "within_5pct": 0.9708402899741192, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008023051076658672, "visible": 0.007721323406725931, "excluded": 0.06508261353718361, "within_5pct": 0.9834581082534862, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3486580913886428, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.015638013370335102, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.010809795842729912, "visible": 0.009626887506661738, "excluded": 0.07113511964045094, "within_5pct": 0.9766481913652275, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007810449157919729, "visible": 0.007303511503063782, "excluded": 0.07364951468001533, "within_5pct": 0.9785482852812014, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3486580913886428, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.015638013370335102, "rotation_degrees": 0.6862415671348572, "center_rmse": 0.019015619531273842, "focal_difference": 0.04241907596588135 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.02573616895850332, "visible": 0.015613560442197675, "excluded": 0.3301892154076803, "within_5pct": 0.9185224854069101, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01005711477998213, "visible": 0.008704866729107466, "excluded": 0.11959342250870592, "within_5pct": 0.9798789134734287, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45227896980941296, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 1.0049808025360107, "center_rmse": 0.018283041194081306, "focal_difference": 0.008793413639068604 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.00588957089193682, "visible": 0.00541324628861121, "excluded": 0.08048081217224408, "within_5pct": 0.998681148795104, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0038363158368551406, "visible": 0.003643439069620877, "excluded": 0.06360180846426661, "within_5pct": 0.9836242269267365, "coverage": 0.9911749790916993, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5590015109628439, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.009435661137104034, "rotation_degrees": 0.3870031237602234, "center_rmse": 0.009819205850362778, "focal_difference": 0.005551040172576904 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0020231983946647934, "pair_records": 18 }, "background": { "discrepancy": 0.0013386593721172724, "pair_records": 18 } }, "reference_discrepancy": 0.006453599780797958, "inference_seconds": 27.172987970523536, "peak_allocated_gib": 8.76651906967163, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0020321131817658366, "pair_records": 18 }, "background": { "discrepancy": 0.0013755386925575733, "pair_records": 18 } }, "reference_discrepancy": 0.006650441559031606, "inference_seconds": 27.71986404620111, "peak_allocated_gib": 8.650798320770264, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.002255190361308428, "pair_records": 18 }, "background": { "discrepancy": 0.0013973563430313774, "pair_records": 18 } }, "reference_discrepancy": 0.006805767305195332, "inference_seconds": 24.643542506732047, "peak_allocated_gib": 8.650798320770264, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.004048520051680471, "visible": 0.003836140536338185, "excluded": 0.06695801959506172, "within_5pct": 0.9906174119483667, "coverage": 0.2607185714054502, "pair_records": 18 }, "background": { "discrepancy": 0.0023551153864692274, "visible": 0.0023551153864692274, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.001487825115066187, "pair_records": 7 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "wineglass/160_17577_32970": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.02081634497292887, "visible": 0.010342471809417414, "excluded": 0.25925499345686853, "within_5pct": 0.9485403949657367, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004320197099797079, "visible": 0.003830341951179264, "excluded": 0.06321898657823728, "within_5pct": 0.9804604963535849, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4629630483686924, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.03164433129131794, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.018299205602054263, "visible": 0.012677759280704269, "excluded": 0.22297120569531362, "within_5pct": 0.8720166677353814, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.03716919786201094, "visible": 0.02468454806292808, "excluded": 0.4249227099555616, "within_5pct": 0.7796787617368446, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6983392676338553, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.01944265328347683, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.013051750087240924, "visible": 0.007555896989185821, "excluded": 0.2107723252701328, "within_5pct": 0.8912509822002675, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0019234539658510492, "visible": 0.001842264880350057, "excluded": 0.03477418893351837, "within_5pct": 0.9920394379815007, "coverage": 0.9927529783826026, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6186733581125736, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.010287392418831587, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.017587826063013193, "visible": 0.012825770599799047, "excluded": 0.1570102149808462, "within_5pct": 0.7672903067216562, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.008875683856980034, "visible": 0.007604323835115926, "excluded": 0.06507409173741734, "within_5pct": 0.9842405938800045, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.8336115404963493, "peak_allocated_gib": 3.3874611854553223, "reference_discrepancy": 0.014143134001642466, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.025181873534591816, "visible": 0.01609638188526785, "excluded": 0.24931262312099417, "within_5pct": 0.8441452022709685, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007167774160384411, "visible": 0.0068234432313259, "excluded": 0.05272754480156715, "within_5pct": 0.9852761720494075, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3489300487563014, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.024529485031962395, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.021904770163331685, "visible": 0.01212393169295224, "excluded": 0.2596426955915217, "within_5pct": 0.8536275772682351, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006961573318104075, "visible": 0.006515852303261094, "excluded": 0.04974540083219231, "within_5pct": 0.9852327280666615, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3489300487563014, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.024529485031962395, "rotation_degrees": 0.7199791669845581, "center_rmse": 0.01252806931734085, "focal_difference": 0.016009092330932617 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.02340333884793555, "visible": 0.011435580619509627, "excluded": 0.25000471520971257, "within_5pct": 0.9096644315095905, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0045453119660341604, "visible": 0.0038434488574595766, "excluded": 0.10661978335424016, "within_5pct": 0.9549251775306048, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.9984376030042768, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 0.5026258230209351, "center_rmse": 0.027689488604664803, "focal_difference": 0.0568508505821228 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.014243439023339341, "visible": 0.00960591006138909, "excluded": 0.1905642998202022, "within_5pct": 0.8710376606167634, "coverage": 0.9981502600329919, "pair_records": 18 }, "background": { "discrepancy": 0.0026197590172390563, "visible": 0.002518776901128149, "excluded": 0.03253233735647914, "within_5pct": 0.9939635844876323, "coverage": 0.9919806747514975, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5584417339414358, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.01572998147457838, "rotation_degrees": 0.32196420431137085, "center_rmse": 0.01140936091542244, "focal_difference": 0.006633102893829346 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.005100984256370427, "pair_records": 18 }, "background": { "discrepancy": 0.0011920177299059307, "pair_records": 18 } }, "reference_discrepancy": 0.008784186094999313, "inference_seconds": 21.843188273720443, "peak_allocated_gib": 8.770070552825928, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.004867948875440578, "pair_records": 18 }, "background": { "discrepancy": 0.0011292673330055665, "pair_records": 18 } }, "reference_discrepancy": 0.009339716285467148, "inference_seconds": 21.380460628308356, "peak_allocated_gib": 8.651086330413818, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.00503111448834229, "pair_records": 18 }, "background": { "discrepancy": 0.0011117442561268474, "pair_records": 18 } }, "reference_discrepancy": 0.009114943444728851, "inference_seconds": 17.656976794824004, "peak_allocated_gib": 8.651086330413818, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.005694560563700798, "visible": 0.0045943501069148265, "excluded": 0.1388601676087908, "within_5pct": 0.9513992343317401, "coverage": 0.543994576622214, "pair_records": 18 }, "background": { "discrepancy": 0.0022639111426708035, "visible": 0.002254065971950292, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0026058149678907035, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "laptop/457_64638_127324": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005049424637321803, "visible": 0.004697813188560917, "excluded": 0.005048389578761441, "within_5pct": 0.998249541821652, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.00507008338794329, "visible": 0.004938669631135439, "excluded": 0.013079529241913315, "within_5pct": 0.9966061761404899, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.46438166964799166, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.0072569092735648155, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0028661029513105795, "visible": 0.002775383475982653, "excluded": 0.001409966758029025, "within_5pct": 0.9999455040871934, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.014480260631310847, "visible": 0.012367848218360204, "excluded": 0.05936007144081855, "within_5pct": 0.9947437431989001, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6802891679108143, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.004233754822053015, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0024900465110843997, "visible": 0.0023984777000019336, "excluded": 0.00012643590381364441, "within_5pct": 1.0, "coverage": 0.9911381149578948, "pair_records": 18 }, "background": { "discrepancy": 0.0019692927942621943, "visible": 0.0019026459271106059, "excluded": 0.009551490392469865, "within_5pct": 0.997327080400912, "coverage": 0.9763548673737177, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6182691371068358, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.0029745500069111586, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.009924301180734047, "visible": 0.008562749740598185, "excluded": 0.09922640330595225, "within_5pct": 0.9847177473636975, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007319433255670487, "visible": 0.007003920788771519, "excluded": 0.03410408745247148, "within_5pct": 0.9966474194167756, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7565286038443446, "peak_allocated_gib": 3.381706714630127, "reference_discrepancy": 0.006654833210632205, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.02210297592201046, "visible": 0.016999288350558722, "excluded": 0.015528890521001626, "within_5pct": 0.9984721691450207, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.02031442651479912, "visible": 0.016667813956883083, "excluded": 0.010045124502451638, "within_5pct": 0.9954907901404526, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34973841719329357, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.017546663992106915, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005472894816369039, "visible": 0.005453882290480081, "excluded": 0.0038206396548802523, "within_5pct": 0.9989754557369808, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.007453980037074511, "visible": 0.0071895888832124895, "excluded": 0.009286061481177277, "within_5pct": 0.996535328092885, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34973841719329357, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.017546663992106915, "rotation_degrees": 0.49967455863952637, "center_rmse": 0.020487479865550995, "focal_difference": 0.026659905910491943 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.00617119603321848, "visible": 0.0060378011306183536, "excluded": 0.011455757392604385, "within_5pct": 0.9988577689022157, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0061027218217917095, "visible": 0.005680220816510579, "excluded": 0.020747299213872326, "within_5pct": 0.9951435796685488, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4516312927007675, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 2.563004732131958, "center_rmse": 0.05815434083342552, "focal_difference": 0.20488649606704712 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.0015186646761866027, "visible": 0.0014734242440855227, "excluded": 0.00034488899460582794, "within_5pct": 1.0, "coverage": 0.9787876651709603, "pair_records": 18 }, "background": { "discrepancy": 0.0018489627409770041, "visible": 0.0018349612983661156, "excluded": 0.009463628517500402, "within_5pct": 0.9983724739508424, "coverage": 0.9765423331881122, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5608585812151432, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.009571266360580921, "rotation_degrees": 0.6225878000259399, "center_rmse": 0.019902938976883888, "focal_difference": 0.01296311616897583 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0007824422287057362, "pair_records": 18 }, "background": { "discrepancy": 0.0007386218144182919, "pair_records": 18 } }, "reference_discrepancy": 0.0036564504262059927, "inference_seconds": 25.25028586294502, "peak_allocated_gib": 8.766873836517334, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0007704889132216888, "pair_records": 18 }, "background": { "discrepancy": 0.0007292820996601904, "pair_records": 18 } }, "reference_discrepancy": 0.0038859773194417357, "inference_seconds": 26.014583373442292, "peak_allocated_gib": 8.652263164520264, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0007524200403392344, "pair_records": 18 }, "background": { "discrepancy": 0.0007633467082878297, "pair_records": 18 } }, "reference_discrepancy": 0.003887534374371171, "inference_seconds": 22.75136628653854, "peak_allocated_gib": 8.652263164520264, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.001735786413748758, "visible": 0.001675314080278652, "excluded": 0.0040487332583122124, "within_5pct": 1.0, "coverage": 0.43826254684245664, "pair_records": 18 }, "background": { "discrepancy": 0.001791474847301067, "visible": 0.0017863391906695518, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0038436467376633983, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "toytrain/350_36700_68417": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006488807013787658, "visible": 0.0052283556799557086, "excluded": 0.16976569455389146, "within_5pct": 0.9830772186488057, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.004949598673963564, "visible": 0.0048156407452985064, "excluded": 0.0096519175005745, "within_5pct": 0.999937280481686, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.46359582897275686, "peak_allocated_gib": 7.615289211273193, "reference_discrepancy": 0.005200405139476061, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.004659209521406993, "visible": 0.002972526464108868, "excluded": 0.21647965027935573, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.015816058942657866, "visible": 0.012764792898315967, "excluded": 0.2256688258019178, "within_5pct": 0.9989753112781427, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6817298429086804, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.004790512379258871, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.004119302020198341, "visible": 0.003454131932935412, "excluded": 0.18190919948545609, "within_5pct": 1.0, "coverage": 0.9986853637160386, "pair_records": 8 }, "background": { "discrepancy": 0.0029809514455818873, "visible": 0.0029801012676045633, "excluded": 0.0011338518930536102, "within_5pct": 0.9998601090008904, "coverage": 0.9940555028232088, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6171894129365683, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.004495673347264528, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006998286531001945, "visible": 0.004793135614274925, "excluded": 0.17459699367003895, "within_5pct": 0.9994314951677089, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.002491830219774456, "visible": 0.0024910148677195687, "excluded": 0.0012622031562036541, "within_5pct": 0.9997221012964423, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7622353369370103, "peak_allocated_gib": 3.3874611854553223, "reference_discrepancy": 0.006253756117075682, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.019881300457926996, "visible": 0.013718401925699363, "excluded": 0.18872403749371214, "within_5pct": 0.990576873372796, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.020440879359173422, "visible": 0.016771091435724316, "excluded": 0.016492139151355962, "within_5pct": 0.9995210836058421, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34882688894867897, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.008081476204097271, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007190636521490806, "visible": 0.005437278128232035, "excluded": 0.1855423046518459, "within_5pct": 0.9955408690887922, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.0069825925686791196, "visible": 0.006842518601719312, "excluded": 0.0057218830334892275, "within_5pct": 0.9999355587060188, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 288, 512 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34882688894867897, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.008081476204097271, "rotation_degrees": 2.3749303817749023, "center_rmse": 0.030223717913031578, "focal_difference": 0.046749770641326904 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.007314878228186191, "visible": 0.006356250111214568, "excluded": 0.12961465764481056, "within_5pct": 0.9701625358698762, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.0040568711496914815, "visible": 0.004054616676350679, "excluded": 0.00045781714205662304, "within_5pct": 0.9999346490654817, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45213720109313726, "peak_allocated_gib": 7.5914740562438965, "reference_discrepancy": null, "rotation_degrees": 2.3000576496124268, "center_rmse": 0.04672737792134285, "focal_difference": 0.027892708778381348 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.004437242028932814, "visible": 0.003640576502335071, "excluded": 0.16292309002371702, "within_5pct": 0.9995764506565015, "coverage": 1.0, "pair_records": 8 }, "background": { "discrepancy": 0.002891782493922278, "visible": 0.002857960238971643, "excluded": 0.0007603834634335829, "within_5pct": 0.9998639482308498, "coverage": 0.9946399992827561, "pair_records": 18 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5657317647710443, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.008465052582323551, "rotation_degrees": 0.39909178018569946, "center_rmse": 0.010291879996657372, "focal_difference": 0.004961878061294556 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.001652194949816208, "pair_records": 8 }, "background": { "discrepancy": 0.0008407063863489809, "pair_records": 18 } }, "reference_discrepancy": 0.0037667411379516125, "inference_seconds": 24.912837533280253, "peak_allocated_gib": 8.76530647277832, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0014934785147273102, "pair_records": 8 }, "background": { "discrepancy": 0.0008600378189417706, "pair_records": 18 } }, "reference_discrepancy": 0.003751276060938835, "inference_seconds": 24.804547604173422, "peak_allocated_gib": 8.652845859527588, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0016094733506214947, "pair_records": 8 }, "background": { "discrepancy": 0.0008170362273241155, "pair_records": 18 } }, "reference_discrepancy": 0.004079759120941162, "inference_seconds": 25.537910817191005, "peak_allocated_gib": 8.652845859527588, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.002552540725518548, "visible": 0.002088648807904347, "excluded": 0.12954601176855182, "within_5pct": 1.0, "coverage": 0.3020198781660789, "pair_records": 8 }, "background": { "discrepancy": null, "visible": null, "excluded": null, "within_5pct": null, "coverage": 0.0001555444638273957, "pair_records": 0 } }, "views": 10, "resolution": [ 294, 518 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "baseballglove/135_15558_31397": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.00790282875786738, "visible": 0.0075508159832279285, "excluded": 0.03681971223093757, "within_5pct": 0.9988643497689781, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005927731682514495, "visible": 0.005906304624889938, "excluded": 0.0032443471661829475, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.46988621819764376, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.018140500411391258, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.007442534962240714, "visible": 0.006864811637729947, "excluded": 0.020536737843848248, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.018363621902366952, "visible": 0.014455058305888557, "excluded": 0.227920441054004, "within_5pct": 0.9999206978588422, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7082256032153964, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.011663791257888079, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0019093812433909067, "visible": 0.0019066682472681758, "excluded": 0.0011988546110047345, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.002755355179031348, "visible": 0.0027133431269043507, "excluded": 0.012432148119724717, "within_5pct": 1.0, "coverage": 0.9924666242626623, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6181768691167235, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.005134259583428502, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.006166313978368695, "visible": 0.005944523119464463, "excluded": 0.006661509156816932, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.009304001134973556, "visible": 0.008105925650099587, "excluded": 0.10617527426628992, "within_5pct": 0.9831259064068111, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7557742986828089, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.006350360112264752, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03470987142824193, "visible": 0.0165529713896812, "excluded": 0.047467570131411486, "within_5pct": 0.999690594059406, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.026010227385784568, "visible": 0.018660529068290695, "excluded": 0.0024757064919409856, "within_5pct": 0.9994823123382226, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.348338863812387, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.040885284543037415, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.003490510387463491, "visible": 0.0034445591574352244, "excluded": 0.008261685201895089, "within_5pct": 0.99947173119876, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.004855311029713521, "visible": 0.004835327376048226, "excluded": 0.0023748427889587684, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.348338863812387, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.040885284543037415, "rotation_degrees": 6.7040815353393555, "center_rmse": 0.11033780872821808, "focal_difference": 0.024309903383255005 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.011130985302494099, "visible": 0.009704583790634717, "excluded": 0.047545061853398844, "within_5pct": 0.994439752422543, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.006870107533093758, "visible": 0.006535707917541961, "excluded": 0.017960977168080157, "within_5pct": 0.9998240022247209, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45160355512052774, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 13.627843856811523, "center_rmse": 0.17670053243637085, "focal_difference": 0.1367136836051941 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.0035844979092852615, "visible": 0.0035673979862269837, "excluded": 0.0011332716349170256, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003022961779454581, "visible": 0.002995713745672584, "excluded": 0.005087624058685152, "within_5pct": 1.0, "coverage": 0.9628492893421454, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5662917187437415, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.04603540152311325, "rotation_degrees": 7.431373596191406, "center_rmse": 0.10640411078929901, "focal_difference": 0.027497470378875732 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0006057319214692617, "pair_records": 18 }, "background": { "discrepancy": 0.0012161686111601894, "pair_records": 18 } }, "reference_discrepancy": 0.04438544064760208, "inference_seconds": 20.390216159634292, "peak_allocated_gib": 8.767762660980225, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.000523446256418052, "pair_records": 18 }, "background": { "discrepancy": 0.0009387818747092237, "pair_records": 18 } }, "reference_discrepancy": 0.040977468714118004, "inference_seconds": 21.77181877940893, "peak_allocated_gib": 8.65080213546753, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0005009955028510573, "pair_records": 18 }, "background": { "discrepancy": 0.001088924895595804, "pair_records": 18 } }, "reference_discrepancy": 0.044045206159353256, "inference_seconds": 21.504240158945322, "peak_allocated_gib": 8.65080213546753, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0020886158847386845, "visible": 0.0020361404759273814, "excluded": 0.006678636058874135, "within_5pct": 1.0, "coverage": 0.5923686195430921, "pair_records": 18 }, "background": { "discrepancy": 0.0016989791065298678, "visible": 0.0016989791065298678, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.0015408887073814132, "pair_records": 10 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "cake/387_46497_92714": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03054945195117672, "visible": 0.014608659480875882, "excluded": 0.4143977221489287, "within_5pct": 0.9257798997580367, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01561959138683958, "visible": 0.012224591754641141, "excluded": 0.1660605136284486, "within_5pct": 0.9865258595661731, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6917828051373363, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.020720619708299637, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03034914089657749, "visible": 0.01505483803340667, "excluded": 0.4473497192891447, "within_5pct": 0.9489910479510739, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.044105199767487765, "visible": 0.018682023890486378, "excluded": 0.3420183982683983, "within_5pct": 0.7350615239025656, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6832743966951966, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.030086373910307884, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.011465396771959452, "visible": 0.0067822663648043815, "excluded": 0.18046889965411264, "within_5pct": 0.9965226778470005, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0073363501328272945, "visible": 0.006100294394863941, "excluded": 0.07645623492694, "within_5pct": 0.979933928853802, "coverage": 0.6263289885871461, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6242621280252934, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.0095383464358747, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.016683501929944326, "visible": 0.010496126779302204, "excluded": 0.29481643941064384, "within_5pct": 0.9815736671929628, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01924298595259058, "visible": 0.014332952832278478, "excluded": 0.19729656115876276, "within_5pct": 0.9324490063071055, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7816585171967745, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.013964866753667593, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.03132346000259404, "visible": 0.01548526015524384, "excluded": 0.4106082362517419, "within_5pct": 0.924474153101021, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01252111244476134, "visible": 0.011465225879042452, "excluded": 0.07714533678416749, "within_5pct": 0.9622127657667598, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3494272558018565, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.031942449510097504, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.021067538683846643, "visible": 0.012431755282193707, "excluded": 0.3738989201228017, "within_5pct": 0.9726833836884553, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.009089814563476621, "visible": 0.007681579387297128, "excluded": 0.04495550804791194, "within_5pct": 0.979525215621994, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3494272558018565, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.031942449510097504, "rotation_degrees": 1.3187553882598877, "center_rmse": 0.027639281004667282, "focal_difference": 0.04894036054611206 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.027568688325876674, "visible": 0.015696036468623612, "excluded": 0.31242364898789476, "within_5pct": 0.8360634432170337, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.021315791000474323, "visible": 0.01570951157270379, "excluded": 0.34045751818560865, "within_5pct": 0.9311065428625154, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45233578886836767, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 31.86712074279785, "center_rmse": 0.6173050999641418, "focal_difference": 0.5489441156387329 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.01348571410663893, "visible": 0.009071498844865994, "excluded": 0.38055012950330824, "within_5pct": 0.9890402557322107, "coverage": 0.9999906610134668, "pair_records": 18 }, "background": { "discrepancy": 0.0097102449802841, "visible": 0.007977111622259331, "excluded": 0.10508058550759149, "within_5pct": 0.9749924481829166, "coverage": 0.611337237178032, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5601244065910578, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.01602631900459528, "rotation_degrees": 1.1385514736175537, "center_rmse": 0.02401822805404663, "focal_difference": 0.01630580425262451 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.005494931160291469, "pair_records": 18 }, "background": { "discrepancy": 0.0031068112331995605, "pair_records": 18 } }, "reference_discrepancy": 0.010017745196819305, "inference_seconds": 20.171214442700148, "peak_allocated_gib": 8.767463207244873, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.006014614867564912, "pair_records": 18 }, "background": { "discrepancy": 0.003538528733835729, "pair_records": 18 } }, "reference_discrepancy": 0.009623204357922077, "inference_seconds": 17.88641467411071, "peak_allocated_gib": 8.649394989013672, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.00598805726321577, "pair_records": 18 }, "background": { "discrepancy": 0.002913940760793368, "pair_records": 18 } }, "reference_discrepancy": 0.00950751407071948, "inference_seconds": 16.124876210466027, "peak_allocated_gib": 8.649394989013672, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.005366120094086393, "visible": 0.0038479097002631444, "excluded": 0.18513424475634066, "within_5pct": 0.9978594752110734, "coverage": 0.537381124417655, "pair_records": 18 }, "background": { "discrepancy": 0.0029237560180456774, "visible": 0.002681210395778521, "excluded": 0.011375376431538919, "within_5pct": 1.0, "coverage": 0.017383062684110655, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ], "skateboard/368_39854_77813": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.008346341783258243, "visible": 0.008039014522558178, "excluded": 0.044810967900881835, "within_5pct": 0.9946999103539335, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005620385707721565, "visible": 0.005615870875796828, "excluded": 0.0011558646799718151, "within_5pct": 1.0, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6518756365403533, "peak_allocated_gib": 7.6137542724609375, "reference_discrepancy": 0.009392369072884321, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0038829908479547388, "visible": 0.003850518956793055, "excluded": 0.029651694578911533, "within_5pct": 0.9977318040339218, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01400733050647258, "visible": 0.01146048419706014, "excluded": 0.1336426543539297, "within_5pct": 0.9968481598893504, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7042793864384294, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.012315579690039158, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.0024270006567616064, "visible": 0.0023412174311294312, "excluded": 0.028344287982240122, "within_5pct": 0.9995566928961807, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.001583455001483959, "visible": 0.0015803119633999818, "excluded": 0.0035535398125848324, "within_5pct": 0.9993742916694061, "coverage": 0.9851316472642715, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6181813105940819, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.0027520827716216445, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.005233428030863202, "visible": 0.004902165377014589, "excluded": 0.05181382349064634, "within_5pct": 0.9982315527257644, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.003682304094274468, "visible": 0.003671745170368831, "excluded": 0.006201361092063087, "within_5pct": 0.9995233011820994, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7773021636530757, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.007971981074661016, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.010818470218191512, "visible": 0.009433019024018593, "excluded": 0.032920402523768225, "within_5pct": 0.999371287056901, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.01153819352930989, "visible": 0.011515317847824943, "excluded": 0.006403852094081447, "within_5pct": 0.9976624458741412, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.3484816960990429, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.013623035978525877, "rotation_degrees": null, "center_rmse": null, "focal_difference": null }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.003997341793041023, "visible": 0.0038817839113182424, "excluded": 0.03276360833779143, "within_5pct": 0.9988754539451965, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.0032264551620153174, "visible": 0.0032202226909037596, "excluded": 0.006041601132431582, "within_5pct": 0.9980557652950872, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 512, 288 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.3484816960990429, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.013623035978525877, "rotation_degrees": 0.7414971590042114, "center_rmse": 0.017219126224517822, "focal_difference": 0.026891648769378662 }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.007076197325024581, "visible": 0.006524881855367972, "excluded": 0.059664803267273275, "within_5pct": 0.9771438124459741, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.005707365888701651, "visible": 0.0055273882711143855, "excluded": 0.0035727478958616716, "within_5pct": 0.9997182483344467, "coverage": 1.0, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.45221580378711224, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 6.350154876708984, "center_rmse": 0.04282332956790924, "focal_difference": 0.20662426948547363 }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.0038021542306118147, "visible": 0.0036866438980217497, "excluded": 0.028999635649477024, "within_5pct": 0.99737569284546, "coverage": 1.0, "pair_records": 18 }, "background": { "discrepancy": 0.002595382851247335, "visible": 0.0025945993114364248, "excluded": 0.002063476968398788, "within_5pct": 0.9993586059432307, "coverage": 0.9807454051697693, "pair_records": 18 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.560017229989171, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.008908369578421116, "rotation_degrees": 0.32793131470680237, "center_rmse": 0.0139744458720088, "focal_difference": 0.010243445634841919 }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.0007263428566493459, "pair_records": 18 }, "background": { "discrepancy": 0.0006005499526878358, "pair_records": 18 } }, "reference_discrepancy": 0.0018657989567145705, "inference_seconds": 25.70559827797115, "peak_allocated_gib": 8.766315937042236, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.0007478156674983997, "pair_records": 18 }, "background": { "discrepancy": 0.0006411930332977605, "pair_records": 18 } }, "reference_discrepancy": 0.0020784190855920315, "inference_seconds": 24.081484037451446, "peak_allocated_gib": 8.652783870697021, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.0007379399878033746, "pair_records": 18 }, "background": { "discrepancy": 0.0006554706770442657, "pair_records": 18 } }, "reference_discrepancy": 0.0018747275462374091, "inference_seconds": 20.612085385248065, "peak_allocated_gib": 8.652783870697021, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X" }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0009874651954598138, "visible": 0.0009873747478514328, "excluded": 0.0022918238615958868, "within_5pct": 1.0, "coverage": 0.7510741605152772, "pair_records": 18 }, "background": { "discrepancy": null, "visible": null, "excluded": null, "within_5pct": null, "coverage": 0.00047186614127769954, "pair_records": 0 } }, "views": 10, "resolution": [ 518, 294 ], "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null } ] }, "aggregate": [ { "method": "map", "label": "MapAnything / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.013953900074271058, "visible": 0.009292199197132815, "excluded": 0.19442131513510727, "within_5pct": 0.9524456426509063, "coverage": 1.0 }, "background": { "discrepancy": 0.005699663715233552, "visible": 0.00552496969998621, "excluded": 0.02740302304147855, "within_5pct": 0.9941744333611822, "coverage": 1.0 } }, "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.4747661747969687, "peak_allocated_gib": 7.614623308181763, "reference_discrepancy": 0.016905834432691336, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "da3", "label": "DA3", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.010545009895677372, "visible": 0.0072970976718277816, "excluded": 0.21540574042682273, "within_5pct": 0.9924692620154428, "coverage": 1.0 }, "background": { "discrepancy": 0.02618071147214292, "visible": 0.015007339012183644, "excluded": 0.2331453007910298, "within_5pct": 0.9713571418567387, "coverage": 1.0 } }, "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.7187953600659966, "peak_allocated_gib": 9.980850219726562, "reference_discrepancy": 0.01434025028720498, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "pi3x", "label": "Pi3X / CO3D inputs", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.004975969439191579, "visible": 0.004434276717965439, "excluded": 0.1508577434900992, "within_5pct": 0.9984543648029833, "coverage": 0.9999019063877044 }, "background": { "discrepancy": 0.0026269443112968615, "visible": 0.0025505552179363685, "excluded": 0.0284585640656497, "within_5pct": 0.9935664274642753, "coverage": 0.9871291548124883 } }, "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.6223410875536501, "peak_allocated_gib": 8.193115234375, "reference_discrepancy": 0.004999430733732879, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "pow3r", "label": "Pow3R (pairs)", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.009430207245165727, "visible": 0.008012657682978527, "excluded": 0.17619874492391577, "within_5pct": 0.9831457072783302, "coverage": 1.0 }, "background": { "discrepancy": 0.008544001475338353, "visible": 0.0075231004955395794, "excluded": 0.06102144435974806, "within_5pct": 0.9822541209882156, "coverage": 1.0 } }, "camera_conditioning": true, "camera_source": "CO3D", "inference_seconds": 0.8241797164082527, "peak_allocated_gib": 3.3854212760925293, "reference_discrepancy": 0.01204023347236216, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "omega", "label": "VGGT-Omega / CO3D cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.02099213818996873, "visible": 0.013804832512975087, "excluded": 0.20040015674643155, "within_5pct": 0.9767310137972537, "coverage": 1.0 }, "background": { "discrepancy": 0.012029652987035615, "visible": 0.011222985896764294, "excluded": 0.03371388665859525, "within_5pct": 0.9910032011037639, "coverage": 1.0 } }, "camera_conditioning": false, "camera_source": "CO3D", "inference_seconds": 0.34919313341379166, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.022003648336976767, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "omega_own", "label": "VGGT-Omega / predicted cameras", "protocol": "common", "regions": { "foreground": { "discrepancy": 0.01070104298997216, "visible": 0.007920387943586673, "excluded": 0.1679343387905164, "within_5pct": 0.9789385643364599, "coverage": 1.0 }, "background": { "discrepancy": 0.005270730282756976, "visible": 0.004980131627559475, "excluded": 0.026322394957776586, "within_5pct": 0.9921848776378787, "coverage": 1.0 } }, "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.34919313341379166, "peak_allocated_gib": 6.35226583480835, "reference_discrepancy": 0.022003648336976767, "rotation_degrees": 0.8740347027778625, "center_rmse": 0.02075232472270727, "focal_difference": 0.03327634930610657, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "map_rgb_only", "label": "MapAnything / RGB only / predicted cameras", "protocol": "map_rgb_only", "regions": { "foreground": { "discrepancy": 0.015276873120417856, "visible": 0.01092122957117724, "excluded": 0.21853460875699782, "within_5pct": 0.9409460439577315, "coverage": 1.0 }, "background": { "discrepancy": 0.006900561588690974, "visible": 0.006507146207154732, "excluded": 0.032508159678629356, "within_5pct": 0.9931178495417826, "coverage": 1.0 } }, "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.4611924481578171, "peak_allocated_gib": 7.593444347381592, "reference_discrepancy": null, "rotation_degrees": 2.4481581449508667, "center_rmse": 0.04359784908592701, "focal_difference": 0.09631195664405823, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "pi3x_rgb_only", "label": "Pi3X / RGB only / predicted cameras", "protocol": "pi3x_rgb_only", "regions": { "foreground": { "discrepancy": 0.005661300453217943, "visible": 0.004497152952604168, "excluded": 0.14255484773368532, "within_5pct": 0.9974075577225912, "coverage": 0.9999953305067334 }, "background": { "discrepancy": 0.0027403758119366097, "visible": 0.002692501032025069, "excluded": 0.02566738439486065, "within_5pct": 0.9943998285347286, "coverage": 0.9847262160176113 } }, "camera_conditioning": false, "camera_source": "predicted", "inference_seconds": 0.5600402508862317, "peak_allocated_gib": 8.075536727905273, "reference_discrepancy": 0.014929486671462655, "rotation_degrees": 0.5579678416252136, "center_rmse": 0.013592119794338942, "focal_difference": 0.011230036616325378, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "ggpt", "label": "Pi3X + GGPT / CO3D inputs", "protocol": "ggpt_native", "regions": { "foreground": { "discrepancy": 0.001574452929724152, "visible": null, "excluded": null, "within_5pct": null, "coverage": null }, "background": { "discrepancy": 0.0009585448679997634, "visible": null, "excluded": null, "within_5pct": null, "coverage": null } }, "reference_discrepancy": 0.007099977694451809, "inference_seconds": 24.94518323848024, "peak_allocated_gib": 8.765811204910278, "camera_conditioning": true, "camera_source": "SfM-refined Pi3X", "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "ggpt_rgb_only", "label": "Pi3X + GGPT / RGB only", "protocol": "ggpt_rgb_only_native", "regions": { "foreground": { "discrepancy": 0.001511577254467007, "visible": null, "excluded": null, "within_5pct": null, "coverage": null }, "background": { "discrepancy": 0.0009217322114221326, "visible": null, "excluded": null, "within_5pct": null, "coverage": null } }, "reference_discrepancy": 0.0068881368497386575, "inference_seconds": 23.379257258959115, "peak_allocated_gib": 8.650096654891968, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X", "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "ggpt_rgb_only_fast", "label": "Pi3X + GGPT / RGB only / RoMaV2-fast", "protocol": "ggpt_rgb_only_fast_native", "regions": { "foreground": { "discrepancy": 0.001589619464589701, "visible": null, "excluded": null, "within_5pct": null, "coverage": null }, "background": { "discrepancy": 0.0009333433601697163, "visible": null, "excluded": null, "within_5pct": null, "coverage": null } }, "reference_discrepancy": 0.006792805623263121, "inference_seconds": 21.546575983054936, "peak_allocated_gib": 8.650096654891968, "camera_conditioning": false, "camera_source": "SfM-refined Pi3X", "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 20 } }, { "method": "co3d_reference", "label": "Original CO3D depths + poses", "protocol": "co3d_reference", "regions": { "foreground": { "discrepancy": 0.0029881592044671244, "visible": 0.0027333209179552405, "excluded": 0.09928470521852416, "within_5pct": 0.9980127762267139, "coverage": 0.53289625646087 }, "background": { "discrepancy": 0.002250635171716587, "visible": 0.002245712586356331, "excluded": 0.0, "within_5pct": 1.0, "coverage": 0.002377583761100437 } }, "camera_conditioning": null, "camera_source": "CO3D", "inference_seconds": null, "peak_allocated_gib": null, "reference_discrepancy": null, "rotation_degrees": null, "center_rmse": null, "focal_difference": null, "scene_count": 20, "available_scenes": 20, "scored_scenes": { "foreground": 20, "background": 18 } } ] }