Research Article
Pillar-Based 3D Object Detection from Point Cloud with Multiattention Mechanism
Table 7
Results on the KITTI test 3D detection benchmark.
| | Model | Car | Pedestrian | Cyclist | | Easy (%) | Mod. (%) | Hard (%) | mAP (%) | Easy (%) | Mod. (%) | Hard (%) | mAP (%) | Easy (%) | Mod. (%) | Hard (%) | mAP (%) |
| | PointPillars | 79.05 | 74.99 | 68.30 | 74.11 | 52.08 | 43.53 | 41.49 | 45.70 | 75.78 | 59.07 | 52.92 | 55.35 | | PointPillars + channel attention | 82.68 | 75.37 | 72.10 | 76.71 | 52.19 | 45.82 | 42.48 | 46.83 | 75.44 | 59.03 | 55.61 | 63.36 | | PointPillars + SOCA | 83.92 | 76.32 | 72.77 | 77.67 | 53.91 | 47.35 | 42.90 | 48.07 | 78.58 | 61.08 | 57.54 | 65.73 |
|
|
