Monocular Depth Estimation on ScanNet++ zero-shot
91.8δ1 AccuracyUniDAC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| UniDACDataset Size=1.45M, Backbone=ViT-L [7], Zero-shot evaluation=true2026.03 | 91.8 | 9.7 | 0.277 | |
| DACUDataset Size=0.79M, Backbone=Swin-L [32], Zero-shot evaluation=true2026.03 | 65.8 | 23.3 | 0.464 | |
| UniK3DDataset Size=7.94M, Backbone=ViT-L [7], Zero-shot evaluation=true2026.03 | 65.1 | 25.3 | 0.285 | |
| Metric3Dv2Dataset Size=16.20M, Backbone=ViT-L [7], Zero-shot evaluation=true2026.03 | 53.6 | 22.3 | 0.895 | |
| UniDepthDataset Size=3.83M, Backbone=ViT-L [7], Zero-shot evaluation=true2026.03 | 36.4 | 49.7 | 1.166 |