Depth Estimation on Matterport3D
96.08delta1HiMODE
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| HiMODE2022.04 | 96.08 | 0.0658 | 0.0245 | 0.3067 | 0.1084 | 99.4 | 99.85 | — | — | |
| NLDPT2022.04 | 95.99 | 0.07 | 0.0287 | 0.3032 | 0.1051 | 99.38 | 99.82 | — | — | |
| DepthAnyDirectionPub' Year=ICLR 20262025.10 | 95.61 | — | — | 0.2882 | — | 98.6 | 99.82 | 0.0912 | 0.0766 | |
| HoHoNet2022.04 | 94.15 | 0.0671 | 0.0417 | 0.3416 | 0.127 | 98.38 | 99.42 | — | — | |
| PAGCNet2025.10 | 93.99 | — | — | 0.2236 | — | 98.52 | 99.84 | 0.0629 | 0.0411 | |
| PanoVGGTInput=Multi-view2026.03 | 92.66 | 0.084 | — | — | — | — | — | — | — | |
| OmniFusionNumber of iterations=22022.03 | 91.89 | 0.09 | 0.0552 | 0.4261 | 0.1483 | 97.97 | 99.31 | — | — | |
| PanoFormer2022.03 | 91.84 | — | — | 0.3635 | — | 98.04 | 99.16 | 0.1013 | 0.0571 | |
| PanoformerPub' Year=ECCV 20222025.10 | 91.84 | — | — | 0.3635 | — | 98.04 | 99.16 | 0.1013 | 0.0571 | |
| DepthMasterIn-domain training status=No, Model specialization=Unified2026.06 | 91.65 | 9.49 | — | 29.85 | — | 97.65 | — | — | — | |
| HRDFusePub' Year=CVPR 20232025.10 | 91.62 | — | — | 0.4433 | — | 96.69 | 98.44 | 0.0967 | 0.0936 | |
| PanoVGGTInput=Monocular2026.03 | 91.57 | 0.0884 | — | — | — | — | — | — | — | |
| EGFormerInput=Monocular2026.03 | 90.82 | 0.0987 | — | — | — | — | — | — | — | |
| GADFNetPub' Year=IEEE TCSVT 20252025.10 | 90.47 | — | — | 0.4564 | — | 96.68 | 98.68 | 0.2592 | 0.0963 | |
| OmniFusionNumber of iterations=12022.03 | 90.4 | 0.098 | 0.0611 | 0.4536 | 0.1587 | 97.57 | 99.19 | — | — | |
| ACDNet2021.12 | 90 | 0.101 | — | 0.4629 | 0.0646 | 96.78 | 98.76 | 0.267 | — | |
| PanoFormerInput=Monocular2026.03 | 89.85 | 0.1007 | — | — | — | — | — | — | — | |
| SvSyn2022.04 | 89.84 | 0.1063 | 0.0599 | 0.4062 | 0.1569 | 97.73 | 99.74 | — | — | |
| SGFormerPub' Year=IEEE TCSVT 20252025.10 | 89.46 | — | — | 0.479 | — | 96.42 | 98.59 | 0.2748 | — | |
| UniFuse with fusion2022.03 | 88.97 | 0.1063 | — | 0.4941 | — | 96.23 | 98.31 | — | — | |
| UniFuse2021.12 | 88.97 | 0.1063 | — | 0.4941 | 0.0701 | 96.23 | 98.31 | 0.2814 | — | |
| UniFuse2022.03 | 88.97 | — | — | 0.4941 | — | 96.23 | 98.31 | 0.2814 | — | |
| BiFuse++Input=Monocular2026.03 | 88.46 | 0.1076 | — | — | — | — | — | — | — | |
| Omnidepth2022.04 | 87.95 | 0.1136 | 0.0691 | 0.4438 | 0.1591 | 97.95 | 99.5 | — | — | |
| HoHoNet2022.03 | 87.86 | 0.1488 | — | 0.5138 | — | 95.19 | 97.71 | — | — | |
| HoHoNet2021.12 | 87.86 | 0.1488 | — | 0.5138 | 0.0871 | 95.19 | 97.71 | 0.2862 | — | |
| SliceNet2021.12 | 87.16 | 0.1764 | — | 0.6133 | 0.1045 | 94.83 | 97.16 | 0.3296 | — | |
| SliceNet2022.03 | 87.16 | — | — | — | — | 94.83 | 97.16 | 0.3296 | 0.1764 | |
| GLPanoDepthPub' Year=IEEE TIP 20242025.10 | 86.41 | — | — | 0.5223 | — | 95.61 | 98.08 | 0.2998 | — | |
| HoHoNetInput=Monocular2026.03 | 85.76 | 0.1217 | — | — | — | — | — | — | — | |
| Elite360DInput=Monocular2026.03 | 85.34 | 0.1263 | — | — | — | — | — | — | — | |
| DepthAnyPanoramaPub' Year=arxiv 20252025.10 | 85.18 | — | — | 0.751 | — | 91.13 | 97.64 | 0.7845 | 0.4456 | |
| BiFuse with fusion2022.03 | 84.52 | 0.2048 | — | 0.6259 | — | 93.19 | 96.32 | — | — | |
| BiFuse2021.12 | 84.52 | 0.2048 | — | 0.6259 | 0.1134 | 93.19 | 96.32 | 0.347 | — | |
| Bifuse2022.03 | 84.52 | — | — | 0.6295 | — | 93.19 | 96.32 | 0.347 | 0.2408 | |
| SphereNetBackbone=ResNet-502021.12 | 84.34 | 0.1258 | — | 0.5212 | 0.0778 | 95.49 | 98.17 | 0.3167 | — | |
| Bifuse2022.04 | 83.81 | 0.139 | 0.1359 | 0.6277 | 0.2079 | 94.44 | 98.15 | — | — | |
| UniFuseInput=Monocular2026.03 | 83.14 | 0.1367 | — | — | — | — | — | — | — | |
| DAPIn-domain training status=No, Model specialization=Unified2026.06 | 81.92 | 17.78 | — | 83.4 | — | 93.78 | — | — | — | |
| EGFormerPub' Year=ICCV 20232025.10 | 81.58 | — | — | 0.6025 | — | 93.9 | 97.35 | 0.1473 | 0.1517 | |
| OmniFusionInput=Monocular2026.03 | 80.86 | 0.1614 | — | — | — | — | — | — | — | |
| DACIBackbone=ResNet101, Zero-shot evaluation=true, Training domain=Indoor2026.03 | 77.3 | 0.156 | — | 0.619 | — | — | — | — | — | |
| FCRN2022.03 | 77.03 | 0.2409 | — | 0.6704 | — | 91.74 | 96.17 | — | — | |
| FCRN2022.03 | 77.03 | — | — | 0.6704 | — | 91.74 | 96.17 | 0.4008 | 0.2409 | |
| UniDACBackbone=ViT-L, Zero-shot evaluation=true, Training domain=Universal2026.03 | 74.5 | 0.175 | — | 0.442 | — | — | — | — | — | |
| DACIBackbone=Swin-L, Zero-shot evaluation=true, Training domain=Indoor2026.03 | 72.3 | 0.179 | — | 0.591 | — | — | — | — | — | |
| UniK3DIn-domain training status=No, Model specialization=Unified2026.06 | 69.74 | 21.74 | — | 54.51 | — | 96.32 | — | — | — | |
| RectNet2022.03 | 68.3 | 0.2901 | — | 0.7643 | — | 87.94 | 94.29 | — | — | |
| OmniDepth2022.03 | 68.3 | — | — | 0.7643 | — | 87.94 | 94.29 | 0.4838 | 0.2901 | |
| DACUBackbone=Swin-L, Zero-shot evaluation=true, Training domain=Universal2026.03 | 66.2 | 0.215 | — | 0.662 | — | — | — | — | — | |
| SN360Pub' Year=IEEE Access 20252025.10 | 44.83 | — | — | 0.9392 | — | — | — | 0.9932 | 0.9808 | |
| DACOBackbone=ResNet101, Zero-shot evaluation=true, Training domain=Outdoor2026.03 | 39 | 0.548 | — | 1.63 | — | — | — | — | — | |
| DACOBackbone=Swin-L, Zero-shot evaluation=true, Training domain=Outdoor2026.03 | 33 | 0.924 | — | 2.164 | — | — | — | — | — | |
| DAM v1Backbone=ViT-L, Zero-shot=true2026.03 | — | — | — | 1.1431 | — | — | — | — | — | |
| DAM v2Backbone=ViT-L, Zero-shot=true2026.03 | — | — | — | 0.5522 | — | — | — | — | — | |
| MarigoldBackbone=SD 2.0 [33], Zero-shot=true2026.03 | — | — | — | 0.5745 | — | — | — | — | — | |
| PanDA-LBackbone=ViT-L, Zero-shot=true2026.03 | — | 0.1036 | — | 0.4539 | — | — | — | — | — | |
| RePer-360Backbone=ViT-L, Zero-shot=true2026.03 | — | 0.1033 | — | 0.4534 | — | — | — | — | — |