nuScenes
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| 3D Object Detection | nuScenes (val) | NDS78.2 | 941 | |
| 3D Object Detection | nuScenes (test) | mAP76.5 | 829 | |
| Semantic Segmentation | nuScenes (val) | mIoU (Segmentation)81.2 | 212 | |
| 3D object detection | nuScenes v1.0 (test) | mAP77.9 | 210 | |
| 3D Object Detection | nuScenes v1.0 (val) | mAP (Overall)71.2 | 190 | |
| LiDAR Semantic Segmentation | nuScenes (val) | mIoU82 | 169 | |
| Open-loop planning | nuScenes (val) | L2 Error (3s)0.41 | 151 | |
| LiDAR Semantic Segmentation | nuScenes official (test) | mIoU84.6 | 132 | |
| 3D Multi-Object Tracking | nuScenes (test) | ID Switches93 | 130 | |
| 3D Multi-Object Tracking | nuScenes (val) | AMOTA76 | 115 | |
| 3D Object Detection | nuScenes v1.0-trainval (val) | NDS72.9 | 87 | |
| Semantic Segmentation | nuScenes (test) | mIoU85.1 | 75 | |
| Vectorized HD Map Construction | nuScenes v1.0-trainval (val) | mAP73.8 | 72 | |
| Vectorized Map Construction | nuScenes v1.0 (val) | AP (Divider)74.4 | 64 | |
| Open-loop planning | nuScenes v1.0 (val) | L2 (1s)0.11 | 59 | |
| Planning | nuScenes (val) | Collision Rate (Avg)0.07 | 52 | |
| Semantic Segmentation | nuScenes v1.0 (test) | mIoU80.5 | 44 | |
| Object Detection | nuScenes (val) | mAP65.5 | 41 | |
| Planning | nuScenes v1.0-trainval (val) | ST-P3 L2 Error (1s)0.11 | 39 | |
| Video Generation | nuScenes (val) | FVD37.23 | 37 | |
| 3D Semantic Segmentation | nuScenes (val) | mIoU81.5 | 37 | |
| Trajectory Prediction | nuScenes Prediction Challenge (test) | minADE (K=10)0.78 | 37 | |
| Multi-agent Scenario Generation | nuScenes (train) | CS Score0.93 | 36 | |
| 3D Semantic Segmentation | nuScenes (test) | mIoU81.9 | 36 | |
| 3D Semantic Segmentation | nuScenes Day to Night (target) | mIoU 3D69.8 | 34 |
Showing 25 of 384 rows
...