Semantic Occupancy Forecasting on SemanticKITTI (val)
9.89mIoU (1s)ForecastOcc-COTR
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| ForecastOcc-COTRinput=monocular camera images, backbone=COTR2026.02 | 9.89 | 7.83 | 6.58 | 8.1 | 35.05 | 32.76 | 30.42 | 32.74 | |
| ForecastOccinput=monocular camera images2026.02 | 9.1 | 7.22 | 6.17 | 7.49 | 33.38 | 31.38 | 29.78 | 31.51 | |
| PDCastinput=monocular camera images2026.02 | 7.48 | 5.72 | 4.58 | 5.92 | 32.44 | 29.51 | 27.65 | 29.86 | |
| Naiveinput=monocular camera images2026.02 | 5.32 | 4.12 | 2.69 | 4.04 | 25.31 | 21.23 | 19.45 | 21.99 |