List your company

Depth Estimation on iBims 1 (test)

0.118RELDMD-MIX
0.112320.150660.1890.22734Jan 19, 2022Jun 1, 2022Oct 12, 2022Feb 22, 2023Jul 5, 2023Nov 15, 2023Mar 27, 2024

Evaluation results

MethodLinks
DMD-MIXtraining_data=NYU, KITTI, Taskonomy and nuScenes, evaluation_mode=zero-shot, depth_constraint=< 10m2023.120.11885.9——0.447———
DMD-NKtraining_data=NYU and KITTI, evaluation_mode=zero-shot, depth_constraint=< 10m2023.120.1380.1——0.563———
Metric3D v2 (Conv-L CSTM_image)Metric Head=Unified, Backbone=ConvNeXt-Large2024.030.144———0.646———
LoLep-64Number of Planes=64, Training Set=RealEstate10K, Inference Batch Size=12023.070.158194990.620.06—3.02
Metric3D v2 (Conv-L CSTM_label)Metric Head=Unified, Backbone=ConvNeXt-Large2024.030.16———0.521———
ECoDepthZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.16368.8——0.664—23.1—
ZoeD-M12-NEvaluation protocol=zero-shot, Relative depth pre-training=M12, Metric depth fine-tuning=NYU, # metric heads=12023.020.16965.894.798.50.7110.083——
ZoeD-M12-NZero-shot transfer=true, Training datasets=12 datasets, Evaluation depth cap=10m2024.030.16965.8——0.711—18.5—
ZoeDepthMetric Head=KITTI and NYU2024.030.169———0.711———
MINE-64Number of Planes=64, Training Set=RealEstate10K, Inference Batch Size=12023.070.177591960.730.08—3.95
ZoeD-NK-NEvaluation protocol=zero-shot, Relative depth pre-training=NYU+KITTI, Metric depth fine-tuning=NYU, # metric heads=12023.020.17267.193.998.30.7350.084——
ZoeD-X-NEvaluation protocol=zero-shot, Relative depth pre-training=None, Metric depth fine-tuning=NYU, # metric heads=12023.020.17366.894.498.30.730.084——
Metric3D v2 (ViT-L CSTM_image)Metric Head=Unified, Backbone=ViT-Large2024.030.185———0.592———
ZoeD-M12-NKEvaluation protocol=zero-shot, Relative depth pre-training=M12, Metric depth fine-tuning=NYU+KITTI, # metric heads=22023.020.18661.592.898.20.7770.092——
ZoeD-M12-NKevaluation_mode=zero-shot, depth_constraint=< 10m2023.120.18661.5——0.777———
VPDZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.18762.7——0.767—11.5—
ACED2022.010.26087951.030.1——
Global-Local Path NetworksTraining Dataset=NYU Depth V2, Evaluation Protocol=Zero-shot / No fine-tuning2022.010.26189961.010.1——
NeWCRFEvaluation protocol=zero-shot2023.020.20654.888.497.90.8610.102——
NeWCRFsevaluation_mode=zero-shot, depth_constraint=< 10m2023.120.20654.8——0.861———
NeWCRFsZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.20654.8——0.861—0—
NewCRFsMetric Head=KITTI or NYU2024.030.206———0.861———
AdaBinsfine-tuning=none2022.030.215587960.90.11——
LocalBinsfine-tuning=none2022.030.215688970.880.1——
LocalBinsEvaluation protocol=zero-shot2023.020.21155.887.796.60.880.104——
LocalBinsevaluation_mode=zero-shot, depth_constraint=< 10m2023.120.21155.8——0.88———
LocalBinsZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.21155.8——0.88—-0.7—
AdaBinsEvaluation protocol=zero-shot2023.020.21255.587.3960.9010.107——
AdaBinsevaluation_mode=zero-shot, depth_constraint=< 10m2023.120.21255.5——0.901———
AdaBinsZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.21255.5——0.901—-2.1—
AdabinsMetric Head=KITTI or NYU2024.030.212———0.901———
AdaBins2022.010.225586951.070.11——
BTSfine-tuning=none2022.030.235486950.930.11——
BTSEvaluation protocol=zero-shot2023.020.23153.886.394.80.9190.112——
BTSevaluation_mode=zero-shot, depth_constraint=< 10m2023.120.23153.8——0.919———
BTSZero-shot transfer=true, Training datasets=NYUv2, Evaluation depth cap=10m2024.030.23153.8——0.919—-6.9—
VNL2022.010.245484931.060.11——
BTS2022.010.245384941.10.12——
DORN2022.010.245581921.130.12——
Metric3D v2 (ViT-g CSTM_image)Metric Head=Unified, Backbone=ViT-giant2024.030.249———0.611———
SharpNet2022.010.265984941.070.11——