4D occupancy forecasting on Occ3D-nuScenes (val)
31.17mIoU (1s)CascadeOcc
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CascadeOccInput=O, FPS=6.00, Memory=13,7842026.06 | 31.17 | 17.91 | 11.94 | 20.34 | 39.72 | 28.6 | 22.5 | 30.28 | |
| OccWorldInput=O, FPS=10.70, Memory=15,7142026.06 | 25.78 | 15.14 | 10.51 | 17.14 | 34.63 | 25.07 | 20.18 | 26.63 | |
| OccLLaMAInput=O&LLM2026.06 | 25.05 | 19.49 | 15.26 | 19.93 | 34.56 | 28.53 | 24.41 | 29.17 | |
| OccLLMInput=O&LLM2026.06 | 24.02 | 21.65 | 17.29 | 20.99 | 36.65 | 32.14 | 28.77 | 32.52 | |
| GEMInput=Camera, with LiDAR init, Aux. Sup.=3D Occ2026.05 | 18.66 | 15.93 | 13.02 | 15.87 | 31.56 | 28.94 | 25.42 | 28.64 | |
| GEMInput=Camera, Aux. Sup.=3D Occ2026.05 | 15.71 | 13.62 | 11.46 | 13.6 | 25.56 | 24.03 | 21.86 | 23.82 | |
| SparseWorldInput=Camera, Aux. Sup.=3D Occ2026.05 | 14.93 | 13.15 | 11.51 | 13.2 | 22.96 | 22.1 | 21.05 | 22.03 | |
| PreWorld + Pre-trainingInput=Camera, Aux. Sup.=2D & 3D Occ2026.05 | 12.27 | 9.24 | 7.15 | 9.55 | 23.62 | 21.76 | 19.63 | 21.62 | |
| PreWorldInput=Camera, Aux. Sup.=3D Occ2026.05 | 11.69 | 8.72 | 6.77 | 9.06 | 23.01 | 20.79 | 18.84 | 20.88 | |
| GWMInput=Camera+LiDAR, Aux. Sup.=3D Occ2026.05 | 11.63 | 10.07 | 8.17 | 10.12 | 26.22 | 24.97 | 22.13 | 24.6 | |
| OccWorld-DInput=Camera, Aux. Sup.=3D Occ2026.05 | 11.55 | 8.1 | 6.22 | 8.62 | 18.9 | 16.26 | 14.43 | 16.53 | |
| OccLLaMA-FInput=Camera+Action+Language, Aux. Sup.=3D Occ2026.05 | 10.34 | 8.66 | 6.98 | 8.66 | 25.81 | 23.19 | 19.97 | 22.99 | |
| OccWorld-TInput=Camera, Aux. Sup.=LiDAR2026.05 | 4.68 | 3.36 | 2.63 | 3.56 | 9.32 | 8.23 | 7.47 | 8.34 |