Depth Estimation on NYU Depth v2
0.119RMSEPi3X
Evaluation results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Pi3XSSI Alignment=504 resolution2026.06 | 0.119 | 98.27 | — | — | — | — | — | — | 0.0341 | — | — | — | — | — | — | — | — | |
| AI PlaygroundTraining Scenario=Brown / Day / High2020.07 | 0.1232 | 59.96 | 84.05 | 93.08 | 0.2835 | 0.1054 | — | — | — | — | — | — | — | — | — | — | — | |
| MoGe-2SSI Alignment=504 resolution2026.06 | 0.127 | 97.58 | — | — | — | — | — | — | 0.0388 | — | — | — | — | — | — | — | — | |
| HYBRIDDEPTHType=DFF2024.07 | 0.128 | 0.995 | 1 | 1 | — | — | — | — | 0.026 | — | — | — | — | — | — | — | — | |
| HYBRIDDEPTHType=DFF2024.07 | 0.128 | 99 | 100 | 100 | — | — | — | — | 0.026 | — | — | — | — | — | — | — | — | |
| WT-STraining Data=3D-asset + RGBD, Inference Steps=50, SSI Alignment=504 resolution2026.06 | 0.1312 | 98.71 | — | — | — | — | — | — | 0.0374 | — | — | — | — | — | — | — | — | |
| WT-STraining Data=3D-asset + RGBD, Inference Steps=20, SSI Alignment=504 resolution2026.06 | 0.1346 | 97.5 | — | — | — | — | — | — | 0.0382 | — | — | — | — | — | — | — | — | |
| MidVCLEpochs=4002025.06 | 0.1354 | — | — | — | 0.2406 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DFVType=DFF2024.07 | 0.136 | 0.996 | 1 | 1 | — | — | — | — | 0.028 | — | — | — | — | — | — | — | — | |
| S-PCLEpochs=4002025.06 | 0.1381 | — | — | — | 0.2428 | — | — | — | — | — | — | — | — | — | — | — | — | |
| WT-STraining Data=3D-asset only, Inference Steps=20, SSI Alignment=504 resolution2026.06 | 0.1384 | 97.35 | — | — | — | — | — | — | 0.0398 | — | — | — | — | — | — | — | — | |
| ShadCLEpochs=4002025.06 | 0.1399 | — | — | — | 0.245 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VGGTSSI Alignment=504 resolution2026.06 | 0.1403 | 97.31 | — | — | — | — | — | — | 0.0415 | — | — | — | — | — | — | — | — | |
| SimCLREpochs=4002025.06 | 0.142 | — | — | — | 0.2545 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReflCLEpochs=4002025.06 | 0.142 | — | — | — | 0.2533 | — | — | — | — | — | — | — | — | — | — | — | — | |
| LaRI-sceneSSI Alignment=504 resolution2026.06 | 0.1424 | 97.08 | — | — | — | — | — | — | 0.048 | — | — | — | — | — | — | — | — | |
| MoCoV2Epochs=4002025.06 | 0.1434 | — | — | — | 0.255 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCLEpochs=4002025.06 | 0.1437 | — | — | — | 0.2587 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOEpochs=4002025.06 | 0.1451 | — | — | — | 0.2595 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DA3SSI Alignment=504 resolution2026.06 | 0.1453 | 96.76 | — | — | — | — | — | — | 0.0458 | — | — | — | — | — | — | — | — | |
| BYOLEpochs=4002025.06 | 0.1458 | — | — | — | 0.2543 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SwAVEpochs=4002025.06 | 0.1459 | — | — | — | 0.2586 | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniDepthV2Evaluation Protocol=zero-shot2026.02 | 0.18 | 98.8 | 99.8 | 100 | — | 0.02 | — | — | 0.046 | — | — | — | — | — | — | — | — | |
| Metric3Dv2Evaluation Protocol=zero-shot2026.02 | 0.183 | 98.9 | 99.8 | 100 | — | 0.02 | — | — | 0.047 | — | — | — | — | — | — | — | — | |
| RAD-LargeEvaluation Protocol=zero-shot, Model Scale=Large2026.02 | 0.192 | 98.8 | 99.8 | 100 | — | 0.02 | — | — | 0.048 | — | — | — | — | — | — | — | — | |
| AI PlaygroundTraining Scenario=Brown / Day / Low2020.07 | 0.201 | 37.2 | 60.62 | 76.27 | 0.501 | 0.1921 | — | — | — | — | — | — | — | — | — | — | — | |
| AI PlaygroundTraining Scenario=Brown / Day / High2020.07 | 0.2014 | 36.66 | 60.12 | 75.86 | 0.5044 | 0.1938 | — | — | — | — | — | — | — | — | — | — | — | |
| Depth AnythingType=SIDE2024.07 | 0.206 | 98 | 99.8 | 100 | — | — | — | — | 0.056 | — | — | — | — | — | — | — | — | |
| DepthAnyV2Evaluation Protocol=fine-tuning2026.02 | 0.206 | 98.4 | 99.8 | 100 | — | 0.024 | — | — | 0.056 | — | — | — | — | — | — | — | — | |
| RAD-BaseEvaluation Protocol=zero-shot, Model Scale=Base2026.02 | 0.211 | 98 | 99.7 | 100 | — | 0.023 | — | — | 0.056 | — | — | — | — | — | — | — | — | |
| ECoDepthType=SIDE2024.07 | 0.218 | 97 | 99.7 | 99.9 | — | — | — | — | 0.059 | — | — | — | — | — | — | — | — | |
| perceptual diffusion modelMethod category=diffusion-based2023.12 | 0.223 | 97.6 | 99.7 | 99.9 | 0.061 | — | — | — | — | — | — | — | — | — | — | — | — | |
| TADPMethod category=diffusion-based2023.12 | 0.225 | 97.6 | 99.7 | 99.9 | 0.062 | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAD-SmallEvaluation Protocol=zero-shot, Model Scale=Small2026.02 | 0.233 | 96.8 | 99.7 | 99.9 | — | 0.028 | — | — | 0.066 | — | — | — | — | — | — | — | — | |
| MeSaMethod category=non-diffusion-based2023.12 | 0.238 | 96.4 | 99.5 | 99.9 | 0.066 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VPDMethod category=diffusion-based2023.12 | 0.254 | 96.4 | 99.5 | 99.9 | 0.069 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VPDType=SIDE2024.07 | 0.254 | 96 | 99.5 | 99.9 | — | — | — | — | 0.069 | — | — | — | — | — | — | — | — | |
| ZoeDepthMethod category=non-diffusion-based2023.12 | 0.27 | 95.5 | 99.5 | 99.9 | 0.075 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ZoeDepthType=SIDE, Version=ZoeDepth-M12-N2024.07 | 0.27 | 96 | 99.5 | 99.9 | — | — | — | — | 0.075 | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-g/14, Evaluation Protocol=DPT2023.04 | 0.279 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ZoeDepthEvaluation Protocol=fine-tuning2026.02 | 0.282 | 95.1 | 99.4 | 99.9 | — | 0.033 | — | — | 0.077 | — | — | — | — | — | — | — | — | |
| SwinV2-MIMMethod category=non-diffusion-based2023.12 | 0.287 | 94.9 | 99.4 | 99.9 | 0.083 | — | — | — | — | — | — | — | — | — | — | — | — | |
| GLIDType=Generalist, Backbone=Swin-L, Data=ImageNet-1K, w/ Labels=false2024.04 | 0.287 | — | — | — | 0.081 | — | — | — | — | — | — | — | — | — | — | — | — | |
| PainterBackbone=ViT-L2022.12 | 0.288 | 95 | — | — | — | — | — | — | 0.08 | — | — | — | — | — | — | — | — | |
| PainterType=Generalist, Backbone=ViT-L, Data=ImageNet-1K, w/ Labels=false2024.04 | 0.288 | — | — | — | 0.08 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-L/14, Evaluation Protocol=DPT2023.04 | 0.293 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GLIDType=Generalist, Backbone=Swin-B, Data=ImageNet-1K, w/ Labels=false2024.04 | 0.293 | — | — | — | 0.085 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-g/14, Evaluation Protocol=lin. 42023.04 | 0.298 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSTM_labelEvaluation protocol=Zero-shot2023.07 | 0.31 | 94.4 | 98.6 | 99.5 | — | 0.035 | — | — | 0.083 | — | — | — | — | — | — | — | — | |
| iDiscEvaluation Protocol=fine-tuning2026.02 | 0.313 | 93.8 | 99.2 | 99.8 | — | 0.037 | — | — | 0.086 | — | — | — | — | — | — | — | — | |
| DepthGenMethod category=diffusion-based2023.12 | 0.314 | 94.6 | 98.7 | 99.6 | 0.074 | — | — | — | — | — | — | — | — | — | — | — | — | |
| IEBinsEvaluation Protocol=fine-tuning2026.02 | 0.314 | 93.6 | 99.2 | 99.8 | — | 0.038 | — | — | 0.087 | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-B/14, Evaluation Protocol=DPT2023.04 | 0.317 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PixelFormerMethod category=non-diffusion-based2023.12 | 0.322 | 92.9 | 99.1 | 99.8 | 0.09 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DDPMethod category=diffusion-based2023.12 | 0.329 | 92.1 | 99 | 99.8 | 0.094 | — | — | — | — | — | — | — | — | — | — | — | — | |
| BinsFormer2022.12 | 0.33 | 92.5 | — | — | — | — | — | — | 0.094 | — | — | — | — | — | — | — | — | |
| Li et al.Note=State-of-the-art reference2023.04 | 0.33 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BinsFormerType=Specialist, Backbone=Swin-L, Data=ImageNet-21K, w/ Labels=false2024.04 | 0.33 | — | — | — | 0.094 | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-L/14, Evaluation Protocol=lin. 42023.04 | 0.333 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NewCRFEvaluation Protocol=fine-tuning2026.02 | 0.333 | 92.1 | 99.1 | 99.8 | — | 0.04 | — | — | 0.095 | — | — | — | — | — | — | — | — | |
| NeWCRFsEvaluation protocol=Supervised2023.07 | 0.334 | 92.2 | 99.2 | 99.8 | — | 0.041 | — | — | 0.095 | — | — | — | — | — | — | — | — | |
| DepthFormerMethod category=non-diffusion-based2023.12 | 0.339 | 92.1 | 98.9 | 99.8 | 0.096 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CSTM_imageEvaluation protocol=Zero-shot2023.07 | 0.341 | 92.5 | 98.3 | 99.4 | — | 0.04 | — | — | 0.092 | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-g/14, Evaluation Protocol=lin. 12023.04 | 0.344 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-S/14, Evaluation Protocol=DPT2023.04 | 0.356 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| iBOTArch=ViT-L/16, Evaluation Protocol=DPT2023.04 | 0.358 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-B/14, Evaluation Protocol=lin. 42023.04 | 0.362 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BinsFormerType=Specialist, Backbone=Swin-B, w/ Labels=false2024.04 | 0.362 | — | — | — | 0.104 | — | — | — | — | — | — | — | — | — | — | — | — | |
| MuRF (Ours)Arch.=ViT-B/14, Evaluation Protocol=lin. 42026.03 | 0.363 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AdabinsEvaluation protocol=Supervised2023.07 | 0.364 | 90.3 | 98.4 | 99.7 | — | 0.044 | — | — | 0.103 | — | — | — | — | — | — | — | — | |
| AdaBinsEvaluation Protocol=fine-tuning2026.02 | 0.364 | 90.3 | 98.4 | 99.7 | — | 0.044 | — | — | 0.103 | — | — | — | — | — | — | — | — | |
| TrDepthEvaluation protocol=Supervised2023.07 | 0.365 | 90 | 98.3 | 99.6 | — | 0.045 | — | — | 0.106 | — | — | — | — | — | — | — | — | |
| MuRF (Ours)Arch.=ViT-B/14, Evaluation Protocol=lin. 12026.03 | 0.368 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Medium ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 42026.03 | 0.372 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| High ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 42026.03 | 0.376 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-L/14, Evaluation Protocol=lin. 12023.04 | 0.384 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Unified-IOBackbone=Unified-IO_XL, Number of Parameters=2925M2022.12 | 0.385 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| iBOTArch=ViT-L/16, Evaluation Protocol=lin. 42023.04 | 0.387 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DepthProEvaluation Protocol=zero-shot2026.02 | 0.387 | 91.8 | 98 | 99.2 | — | 0.045 | — | — | 0.099 | — | — | — | — | — | — | — | — | |
| iBOTArch.=ViT-L/16, Evaluation Protocol=lin. 42026.03 | 0.387 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NYUv2 trained modelTraining Scenario=NYUv2 [1]2020.07 | 0.39 | 89.5 | 98 | 99.6 | 0.103 | 0.043 | — | — | — | — | — | — | — | — | — | — | — | |
| BTSEvaluation Protocol=fine-tuning2026.02 | 0.391 | 88.5 | 97.8 | 99.4 | — | 0.046 | — | — | 0.109 | — | — | — | — | — | — | — | — | |
| Medium ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 12026.03 | 0.394 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| High ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 12026.03 | 0.397 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-B/14, Evaluation Protocol=lin. 12023.04 | 0.399 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Unified-IOType=Generalist, Backbone=Unified-IO-L, Data=95 datasets, w/ Labels=true2024.04 | 0.402 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Low ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 42026.03 | 0.41 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours (DINOv2)Data=INet-22k, Arch.=L/142024.06 | 0.411 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OpenCLIPArch=ViT-G/14, Evaluation Protocol=DPT2023.04 | 0.414 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAEArch=ViT-H/14, Evaluation Protocol=DPT2023.04 | 0.415 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VNLEvaluation protocol=Supervised2023.07 | 0.416 | 87.5 | 97.6 | 99.4 | — | 0.048 | — | — | 0.108 | — | — | — | — | — | — | — | — | |
| iBOTArch=ViT-L/16, Evaluation Protocol=lin. 12023.04 | 0.417 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-S/14, Evaluation Protocol=lin. 42023.04 | 0.417 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| iBOTArch.=ViT-L/16, Evaluation Protocol=lin. 12026.03 | 0.417 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Low ResolutionArch.=ViT-B/14, Evaluation Protocol=lin. 12026.03 | 0.435 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DINOv2Arch=ViT-S/14, Evaluation Protocol=lin. 12023.04 | 0.449 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseDepth2022.12 | 0.465 | 84.6 | — | — | — | — | — | — | 0.123 | — | — | — | — | — | — | — | — | |
| UViMBackbone=ViT-L2022.12 | 0.467 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UVIMType=Generalist, Backbone=ViT-L, Data=ImageNet-21K, w/ Labels=true2024.04 | 0.467 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Unified-IOType=Generalist, Backbone=Unified-IO-B, Data=95 datasets, w/ Labels=true2024.04 | 0.469 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |