Monocular Depth Estimation on NYU-Depth v2 (official)
0.056Abs RelDAR-L
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| DAR-LSize=2.0B, Generative models requiring multiple forward passes=true2026.04 | 0.056 | 0.205 | 0.024 | 98.2 | 99.8 | 100 | 0.011 | |
| DAv1Size=343M, Large-scale pre-training on relative depth=true2026.04 | 0.056 | 0.206 | 0.024 | 98.4 | 99.8 | 100 | — | |
| DAv2Size=343M, Large-scale pre-training on relative depth=true2026.04 | 0.056 | 0.206 | 0.024 | 98.4 | 99.8 | 100 | — | |
| DINOv3-L+LFRSize=350M2026.04 | 0.057 | 0.206 | 0.025 | 98.4 | 99.8 | 100 | 0.017 | |
| DAR-BSize=1.0B, Generative models requiring multiple forward passes=true2026.04 | 0.058 | 0.214 | 0.026 | 98 | 99.7 | 99.9 | 0.013 | |
| ECoDepthVenue=CVPR'242024.03 | 0.059 | 0.218 | 0.026 | 97.8 | 99.7 | 99.9 | 0.013 | |
| EcoDepthSize=954M, Generative models requiring multiple forward passes=true2026.04 | 0.059 | 0.218 | 0.026 | 97.8 | 99.7 | 99.9 | 0.013 | |
| DINOv3-LSize=342M2026.04 | 0.06 | 0.212 | 0.026 | 98.1 | 99.8 | 100 | 0.018 | |
| VPDVenue=ICCV'232024.03 | 0.069 | 0.254 | 0.03 | 96.4 | 99.5 | 99.9 | 0.027 | |
| VPDParams=872M2024.07 | 0.069 | 0.254 | 0.03 | 96.4 | 99.5 | 99.9 | — | |
| VPDSize=600M, Generative models requiring multiple forward passes=true2026.04 | 0.069 | 0.254 | 0.03 | 96.4 | 99.5 | 99.9 | 0.027 | |
| DINOv3-B+LFRSize=127M2026.04 | 0.07 | 0.251 | 0.03 | 96.5 | 99.7 | 99.9 | 0.025 | |
| DINOv3-BSize=120M2026.04 | 0.072 | 0.252 | 0.031 | 96.4 | 99.7 | 99.9 | 0.026 | |
| ScaleDepth-NParams=216M, trained_on=NYU-Depth V22024.07 | 0.074 | 0.267 | 0.032 | 95.7 | 99.4 | 99.9 | — | |
| ZoeDepthVenue=Arxiv'23, Version=ZoeDepth-M12-N2024.03 | 0.075 | 0.27 | 0.032 | 95.5 | 99.5 | 99.9 | 0.03 | |
| ZoeDepthSize=343M, Large-scale pre-training on relative depth=true2026.04 | 0.075 | 0.27 | 0.032 | 95.5 | 99.5 | 99.9 | 0.03 | |
| AiTVenue=ICCV'232024.03 | 0.076 | 0.275 | 0.033 | 95.4 | 99.4 | 99.9 | — | |
| MIMVenue=CVPR'23, Backbone=SwinV2-L 1K2024.03 | 0.083 | 0.287 | 0.035 | 94.9 | 99.4 | 99.9 | — | |
| MIMSize=197M2026.04 | 0.083 | 0.287 | 0.035 | 94.9 | 99.4 | 99.9 | — | |
| DiffusionDepth2024.12 | 0.085 | — | — | 93.9 | — | — | — | |
| iDiscParams=209M2024.07 | 0.086 | 0.313 | 0.037 | 94 | 99.3 | 99.9 | — | |
| NDDepthCap=0-10m, Backbone=Swin-L2023.09 | 0.087 | 0.311 | 0.038 | 93.6 | 99.1 | 99.8 | 0.041 | |
| IEBinsParams=273M2024.07 | 0.087 | 0.314 | 0.038 | 93.6 | 99.2 | 99.8 | — | |
| URCDC-DepthCap=0-10m2023.02 | 0.088 | 0.316 | 0.038 | 93.3 | 99.2 | 99.8 | — | |
| WorDepthBackbone=Swin-L2024.04 | 0.088 | 0.317 | 0.038 | 93.2 | 99.2 | 99.8 | — | |
| PixelFormerVenue=WACV'232024.03 | 0.09 | 0.322 | 0.039 | 92.9 | 99.1 | 99.8 | 0.043 | |
| PixelFormerSize=365M2026.04 | 0.09 | 0.322 | 0.039 | 92.9 | 99.1 | 99.8 | 0.043 | |
| Yu et al.Backbone=Swin-L2024.04 | 0.093 | 0.331 | 0.04 | 92.1 | 99 | 99.8 | — | |
| DDPVenue=ICCV'232024.03 | 0.094 | 0.329 | 0.04 | 92.1 | 99 | 99.8 | — | |
| DepthFormerBackbone=Swin-L2024.04 | 0.094 | 0.329 | 0.04 | 92.3 | 98.9 | 99.7 | — | |
| BinsFormerParams=255M2024.07 | 0.094 | 0.33 | 0.04 | 92.5 | 98.9 | 99.7 | — | |
| BinsFormer2024.12 | 0.094 | — | — | 92.5 | — | — | — | |
| CrossFlow2024.12 | 0.094 | — | — | 92.8 | — | — | — | |
| NeWCRFsCap=0-10m2023.02 | 0.095 | 0.334 | 0.041 | 92.2 | 99.2 | 99.8 | — | |
| NeWCRFsCap=0-10m2023.09 | 0.095 | 0.334 | 0.041 | 92.2 | 99.2 | 99.8 | 0.045 | |
| NeWCRFsVenue=CVPR'222024.03 | 0.095 | 0.334 | 0.041 | 92.2 | 99.2 | 99.8 | 0.045 | |
| NeWCRFsBackbone=Swin-L2024.04 | 0.095 | 0.331 | 0.041 | 92.2 | 99.2 | 99.8 | — | |
| NeWCRFsParams=270M2024.07 | 0.095 | 0.334 | 0.041 | 92.2 | 99.2 | 99.8 | — | |
| NeWCRFsSize=270M2026.04 | 0.095 | 0.334 | 0.041 | 92.2 | 99.2 | 99.8 | 0.045 | |
| DepthFormerCap=0-10m2023.02 | 0.096 | 0.339 | 0.041 | 92.1 | 98.9 | 99.8 | — | |
| DepthFormer2024.12 | 0.096 | — | — | 92.1 | — | — | — | |
| Jun et al.Venue=ECCV'222024.03 | 0.098 | 0.355 | 0.042 | 91.3 | 98.7 | 99.8 | — | |
| BaselineBackbone=Swin-L2024.04 | 0.098 | 0.351 | 0.043 | 91 | 99 | 99.8 | — | |
| LocalbinsVenue=ECCV'222024.03 | 0.099 | 0.357 | 0.042 | 90.7 | 98.7 | 99.8 | — | |
| LocalbinsSize=74M2026.04 | 0.099 | 0.357 | 0.042 | 90.7 | 98.7 | 99.8 | — | |
| Long et al.Cap=0-10m2023.02 | 0.101 | 0.377 | 0.044 | 89 | 98.2 | 99.6 | — | |
| Long et al.Cap=0-10m2023.09 | 0.101 | 0.377 | 0.044 | 89 | 98.2 | 99.6 | — | |
| ASNBackbone=HRNet-482024.04 | 0.101 | 0.377 | 0.044 | 89 | 98.2 | 99.6 | — | |
| AdabinsCap=0-10m2023.02 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| AdabinsCap=0-10m2023.09 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| AdaBinsVenue=CVPR'212024.03 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| ASTransformerBackbone=ViT-B2024.04 | 0.103 | 0.374 | 0.044 | 90.2 | 98.5 | 99.7 | — | |
| AdaBinsBackbone=EffNet-B5 + ViT-mini2024.04 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| AdaBinsParams=78M2024.07 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| AdaBins2024.12 | 0.103 | — | — | 90.3 | — | — | — | |
| AdaBinsSize=40M2026.04 | 0.103 | 0.364 | 0.044 | 90.3 | 98.4 | 99.7 | — | |
| P3DepthCap=0-10m2023.02 | 0.104 | 0.356 | 0.043 | 89.8 | 98.1 | 99.6 | — | |
| P3DepthCap=0-10m2023.09 | 0.104 | 0.356 | 0.043 | 89.8 | 98.1 | 99.6 | — | |
| P3DepthVenue=CVPR'222024.03 | 0.104 | 0.356 | 0.043 | 89.8 | 98.1 | 99.6 | — | |
| P3DepthSize=43M2026.04 | 0.104 | 0.356 | 0.043 | 89.8 | 98.1 | 99.6 | — | |
| PWACap=0-10m2023.02 | 0.105 | 0.374 | 0.045 | 89.2 | 98.5 | 99.7 | — | |
| PWACap=0-10m2023.09 | 0.105 | 0.374 | 0.045 | 89.2 | 98.5 | 99.7 | — | |
| TransDepthCap=0-10m2023.02 | 0.106 | 0.365 | 0.045 | 90 | 98.3 | 99.6 | — | |
| TransDepthCap=0-10m2023.09 | 0.106 | 0.365 | 0.045 | 90 | 98.3 | 99.6 | — | |
| TransDepthBackbone=ViT-B2024.04 | 0.106 | 0.365 | 0.045 | 90 | 98.3 | 99.6 | — | |
| TransDepth2024.12 | 0.106 | — | — | 90 | — | — | — | |
| VNLCap=0-10m2023.02 | 0.108 | 0.416 | 0.048 | 87.5 | 97.6 | 99.4 | — | |
| DAVCap=0-10m2023.02 | 0.108 | 0.412 | — | 88.2 | 98 | 99.6 | — | |
| VNLCap=0-10m2023.09 | 0.108 | 0.416 | 0.048 | 87.5 | 97.6 | 99.4 | — | |
| DAVCap=0-10m2023.09 | 0.108 | 0.412 | — | 88.2 | 98 | 99.6 | — | |
| Yin et al.Backbone=ResNeXt-1012024.04 | 0.108 | 0.416 | 0.048 | 87.5 | 97.6 | 99.4 | — | |
| BTSVenue=Arxiv'192024.03 | 0.11 | 0.392 | 0.047 | 88.5 | 97.8 | 99.4 | 0.066 | |
| DPTVenue=ICCV'212024.03 | 0.11 | 0.357 | 0.045 | 90.4 | 98.8 | 99.8 | — | |
| Big to SmallBackbone=DenseNet-1612024.04 | 0.11 | 0.392 | 0.047 | 88.5 | 97.8 | 99.4 | — | |
| DPT-HybridBackbone=ViT-B2024.04 | 0.11 | 0.357 | 0.045 | 90.4 | 98.8 | 99.8 | — | |
| BTSParams=47M2024.07 | 0.11 | 0.392 | 0.047 | 88.5 | 97.8 | 99.4 | — | |
| BTSSize=20M2026.04 | 0.11 | 0.392 | 0.047 | 88.5 | 97.8 | 99.4 | 0.066 | |
| DPTSize=343M2026.04 | 0.11 | 0.357 | 0.045 | 90.4 | 98.8 | 99.8 | — | |
| Chen et al.Venue=IJCAI-192024.03 | 0.111 | 0.514 | 0.048 | 87.8 | 97.7 | 99.4 | — | |
| SwinV2-LVenue=CVPR'222024.03 | 0.112 | 0.381 | 0.051 | 88.6 | 98.4 | 99.7 | — | |
| SwinV2-LSize=197M2026.04 | 0.112 | 0.381 | 0.051 | 88.6 | 98.4 | 99.7 | — | |
| BTSCap=0-10m2023.02 | 0.113 | 0.407 | 0.049 | 87.1 | 97.7 | 99.5 | — | |
| BTSCap=0-10m2023.09 | 0.113 | 0.407 | 0.049 | 87.1 | 97.7 | 99.5 | 0.066 | |
| Fu et al.Cap=0-10m2023.02 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| Fu et al.Cap=0-10m2023.09 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| DORNVenue=CVPR'182024.03 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| DORNBackbone=ResNet-1012024.04 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| DORNParams=110M2024.07 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| DORNSize=45M2026.04 | 0.115 | 0.509 | 0.051 | 82.8 | 96.5 | 99.2 | — | |
| GeoNetBackbone=ResNet-502024.04 | 0.128 | 0.569 | 0.057 | 83.4 | 96 | 99 | — | |
| SwinV2-BVenue=CVPR'222024.03 | 0.133 | 0.462 | 0.059 | 81.9 | 97.5 | 99.5 | — | |
| SharpNetVenue=ICCV'192024.03 | 0.139 | 0.502 | 0.047 | 83.6 | 96.6 | 99.3 | — | |
| Eigen et al.Cap=0-10m2023.02 | 0.158 | 0.641 | — | 76.9 | 95 | 98.8 | — | |
| Eigen et al.Venue=NIPS'142024.03 | 0.158 | 0.641 | — | 76.9 | 95 | 98.8 | — | |
| Eigen et al.Size=45M2026.04 | 0.158 | 0.641 | — | 76.9 | 95 | 98.8 | — | |
| CLIPMDEBackbone=CLIP2024.04 | 0.319 | 0.97 | 0.139 | 46.5 | 77.6 | 92.2 | — | |
| DepthCLIPBackbone=CLIP (zero-shot)2024.04 | 0.388 | 1.167 | 0.156 | 39.4 | 68.3 | 85.1 | — |