Semantic occupancy prediction on PanoMMOcc 1.0 (test)
23.34mIoUVoxelHound
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VoxelHoundModality=C+L+T+P2026.03 | 23.34 | 26.69 | 21.77 | 0 | 49.53 | 34.97 | 8.59 | 24.61 | 34.41 | 37.35 | 18.67 | 4.71 | 18.76 | |
| VoxelHoundModality=C+L2026.03 | 22.87 | 26.86 | 22.11 | 0 | 48.99 | 34.56 | 6.46 | 23.38 | 34.16 | 37.7 | 16.49 | 7.14 | 16.54 | |
| EFFOcc-TModality=C+L2026.03 | 19.18 | 19.81 | 12.13 | 0.68 | 44.16 | 30.2 | 7.14 | 21.49 | 30.1 | 33.87 | 15.18 | 5.41 | 10 | |
| EFFOcc-LModality=L2026.03 | 18.77 | 20.25 | 12.5 | 1.28 | 44.02 | 29.47 | 8.52 | 19.25 | 28.31 | 31.79 | 16.69 | 4.55 | 8.59 | |
| MonoSceneModality=C2026.03 | 8.94 | 10.56 | 16.02 | 0 | 26.81 | 20.45 | 2.67 | 10.06 | 10.67 | 4.26 | 2.92 | 0.85 | 1.98 | |
| VoxelHoundModality=C+T+P2026.03 | 6.14 | 1.51 | 8.25 | 0.03 | 20.83 | 18.17 | 4.47 | 7.78 | 8.96 | 0.08 | 3.26 | 0.08 | 0.21 | |
| VoxelHoundModality=C2026.03 | 5.79 | 2.88 | 5.06 | 0 | 21.97 | 16.09 | 1.45 | 10.31 | 8.77 | 1.16 | 1.05 | 0.15 | 0.57 | |
| M-CONetModality=C+L2026.03 | 4.68 | 4.95 | 10.16 | 0 | 20.91 | 19.09 | 2.25 | 9.65 | 5.52 | 0.96 | 1.39 | 0.06 | 0.01 | |
| EFFOcc-CModality=C2026.03 | 4.47 | 1.68 | 7.11 | 0 | 19.9 | 15.53 | 0.82 | 5.3 | 2.8 | 0.04 | 0.47 | 0 | 0.01 | |
| C-CONetModality=C2026.03 | 3.79 | 0.47 | 5.13 | 0 | 22.26 | 16.32 | 1.86 | 7.97 | 4.93 | 0.2 | 0.65 | 0.53 | 0.3 | |
| L-CONetModality=L2026.03 | 3.75 | 0.04 | 6.42 | 0 | 21.16 | 16.31 | 1.91 | 6.58 | 5.34 | 0.78 | 1.41 | 0 | 0 |