Video Depth Estimation on KITTI (test)
98.6Delta1pi^3
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| pi^3streaming=false2025.12 | 98.6 | — | — | 0.038 | — | |
| LASER (pi^3)streaming=true, backbone=pi^32025.12 | 98.3 | — | — | 0.054 | — | |
| NVDS+Type=Learning Based, Backbone=DPT-Large2023.07 | 98.2 | 0.046 | 23.3 | — | — | |
| MAMOType=Learning Based2023.07 | 97.7 | 0.049 | — | — | — | |
| DeepV2DType=Learning Based2023.07 | 97.2 | 0.051 | 42.8 | — | — | |
| NVDS+Type=Learning Based, Backbone=MiDaS-v2.1-Large2023.07 | 97 | 0.049 | 25.8 | — | — | |
| DPT-LargeType=Single Image2023.07 | 96.4 | 0.069 | 58.5 | — | — | |
| VGGTstreaming=false2025.12 | 96.3 | — | — | 0.073 | — | |
| WinT3Rstreaming=true2025.12 | 94.9 | — | — | 0.081 | — | |
| STream3R-betastreaming=true2025.12 | 94.7 | — | — | 0.08 | — | |
| MiDaS-v2.1-LargeType=Single Image2023.07 | 94 | 0.088 | 60.2 | — | — | |
| VITAType=Learning Based2023.07 | 91.2 | 0.095 | 31.6 | — | — | |
| TTT3Rstreaming=true2025.12 | 90.4 | — | — | 0.113 | — | |
| Robust-CVDType=Test-time Training2023.07 | 90.1 | 0.097 | 33.8 | — | — | |
| ST-CLSTMType=Learning Based2023.07 | 89 | 0.101 | 41.3 | — | — | |
| FMNetType=Learning Based2023.07 | 88.6 | 0.099 | 37.5 | — | — | |
| LASER (VGGT)streaming=true, backbone=VGGT2025.12 | 88.4 | — | — | 0.116 | — | |
| CUT3Rstreaming=true2025.12 | 88.1 | — | — | 0.118 | — | |
| CVDType=Test-time Training2023.07 | 87.8 | 0.114 | 37.4 | — | — | |
| Cao et al.Type=Learning Based2023.07 | 87.2 | 0.109 | — | — | — | |
| Point3Rstreaming=true2025.12 | 84.2 | — | — | 0.136 | — | |
| VGGT-SLAMstreaming=true2025.12 | 81.8 | — | — | 0.136 | — | |
| WSVDType=Learning Based2023.07 | 81.2 | 0.156 | 49.7 | — | — | |
| Spann3Rstreaming=true2025.12 | 73.7 | — | — | 0.198 | — | |
| StreamVGGTstreaming=true2025.12 | 72.1 | — | — | 0.173 | — | |
| Any4DEvaluation Protocol=Single-Step Feed-Forward2025.12 | — | 0.09 | — | — | 93.97 | |
| CUT3REvaluation Protocol=Single-Step Feed-Forward2025.12 | — | 0.1 | — | — | 89.9 | |
| DepthCrafterEvaluation Protocol=Video Depth2025.12 | — | 0.11 | — | — | 88.5 | |
| DUSt3REvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | — | 0.12 | — | — | 84.9 | |
| MapAnythingEvaluation Protocol=Single-Step Feed-Forward2025.12 | — | 0.09 | — | — | 94.26 | |
| MegaSAMEvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | — | 0.07 | — | — | 91.6 | |
| MonST3REvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | — | 0.08 | — | — | 93.4 | |
| SelfEvo (VGGT)Align=scale2026.04 | — | — | — | 0.047 | 98.1 | |
| SelfEvo (VGGT)Align=scale&shift2026.04 | — | — | — | 0.042 | 97.9 | |
| SpatialTrackerV2Evaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | — | 0.05 | — | — | 97.3 | |
| VDAEvaluation Protocol=Video Depth2025.12 | — | 0.08 | — | — | 95.1 | |
| VGGTEvaluation Protocol=Single-Step Feed-Forward2025.12 | — | 0.09 | — | — | 94.37 | |
| VGGTAlign=scale2026.04 | — | — | — | 0.074 | 96 | |
| VGGTAlign=scale&shift2026.04 | — | — | — | 0.059 | 96.1 |