3D Object Detection on Waymo Open Dataset Vehicle LEVEL_1 v1.2 (test)
77.663D mAP OverallM3DETR
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| M3DETRinput=single point cloud frame2021.04 | 77.66 | 92.54 | 76.27 | 57.12 | 77.09 | 92.09 | 75.61 | 56.02 | |
| PV-RCNN*reproducibility=reproduced results, input=single point cloud frame2021.04 | 76.89 | 92.27 | 75.51 | 55.35 | 76.3 | 91.82 | 74.79 | 54.27 | |
| RangeDet2021.04 | 75.83 | 88.41 | 73.83 | 55.31 | 75.38 | 87.95 | 73.38 | 54.84 | |
| Det3D2021.04 | 73.29 | 90.31 | 70.54 | 49.1 | 72.27 | 89.65 | 68.96 | 47.45 | |
| RCD2021.04 | 69.59 | 87.2 | 67.8 | 46.1 | — | — | — | — | |
| PointPillars2021.04 | 68.62 | 87.2 | 65.5 | 40.92 | 68.08 | 86.71 | 64.87 | 40.19 | |
| StarNet2021.04 | 61.5 | 82.2 | 56.6 | 32.2 | 61 | 81.7 | 56 | 31.8 |