NYU
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Depth Estimation | NYU v2 (test) | Threshold Accuracy (delta < 1.25)98.4 | 423 | |
| Monocular Depth Estimation | NYU v2 (test) | Abs Rel0.049 | 257 | |
| Semantic Segmentation | NYU v2 (test) | mIoU59.13 | 248 | |
| Surface Normal Estimation | NYU v2 (test) | Mean Angle Distance (MAD)8.6 | 206 | |
| Depth Super-Resolution | NYU v2 (test) | RMSE1.49 | 126 | |
| Monocular Depth Estimation | NYU v2 | Delta 1 Acc98.5 | 113 | |
| Surface Normal Prediction | NYU v2 | Mean Error13.34 | 100 | |
| 3D Hand Pose Estimation | NYU (test) | Mean Error (mm)7.06 | 100 | |
| Semantic Segmentation | NYU V2 | mIoU63.6 | 74 | |
| Semantic Scene Completion | NYU v2 (test) | Ceiling Error0 | 72 | |
| Depth Estimation | NYU v2 (val) | RMSE0.201 | 53 | |
| Scene Completion | NYU dataset (test) | mIoU75 | 50 | |
| Scene Completion | NYU v2 (test) | mIoU78.2 | 48 | |
| Semantic Scene Completion | NYU (test) | Ceiling Error0 | 46 | |
| Depth Completion | NYU v2 (val) | RMSE0.09 | 41 | |
| Semantic Segmentation | NYU v2 (val) | mIoU60 | 37 | |
| Depth Super-Resolution / Completion | NYU v2 (test) | AbsRel1.53 | 36 | |
| Depth Super-Resolution | NYU v2 | RMSE0.1197 | 35 | |
| Multi-Task Learning | NYU v2 (test) | Delta m%403 | 31 | |
| Depth Map Super-Resolution | NYU v2 (test) | Value Errors0.04 | 28 | |
| Hand Pose Estimation | NYU (test) | 3D Error (mm)8.29 | 25 | |
| Surface Normal Estimation | NYU v2 | RMSE21.9 | 23 | |
| Monocular Depth Estimation | NYU | AbsRel4.4 | 21 | |
| Single-view depth estimation | NYU official 654 images v2 (test) | AbsRel0.108 | 21 | |
| Single-view depth estimation | NYUv2 36 (test) | AbsRel0.108 | 21 |
Showing 25 of 130 rows