Video Depth Estimation on Average (Bonn, KITTI, Sintel) (test)
0.09Relative ErrorSpatialTrackerV2
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SpatialTrackerV2Evaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | 0.09 | 88.8 | |
| MegaSAMEvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | 0.1 | 87.97 | |
| VGGTEvaluation Protocol=Single-Step Feed-Forward2025.12 | 0.13 | 85.85 | |
| Any4DEvaluation Protocol=Single-Step Feed-Forward2025.12 | 0.13 | 86.28 | |
| MapAnythingEvaluation Protocol=Single-Step Feed-Forward2025.12 | 0.14 | 84.97 | |
| DepthCrafterEvaluation Protocol=Video Depth2025.12 | 0.15 | 85.23 | |
| MonST3REvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | 0.16 | 82.73 | |
| VDAEvaluation Protocol=Video Depth2025.12 | 0.17 | 86.9 | |
| CUT3REvaluation Protocol=Single-Step Feed-Forward2025.12 | 0.21 | 80.3 | |
| DUSt3REvaluation Protocol=Feed-Forward + Iterative Optimization2025.12 | 0.26 | 75.83 |