3D Object Detection on V2XSet Noisy setting
83.6AP@0.5V2X-ViT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| V2X-ViTBackbone=PointPillar, Voxel resolution=0.4 m, Compression rate=32, Fusion strategy=Intermediate, Fine-tuned=true, Encoder layers=3, MSwin window sizes=4, 8, 162022.03 | 83.6 | 61.4 | |
| DiscoNetBackbone=PointPillar, Voxel resolution=0.4 m, Compression rate=32, Fusion strategy=Intermediate2022.03 | 79.8 | 54.1 | |
| V2VNetBackbone=PointPillar, Voxel resolution=0.4 m, Compression rate=32, Fusion strategy=Intermediate2022.03 | 79.1 | 49.3 | |
| Early FusionBackbone=PointPillar, Voxel resolution=0.4 m2022.03 | 72 | 38.4 | |
| F-CooperBackbone=PointPillar, Voxel resolution=0.4 m, Compression rate=32, Fusion strategy=Intermediate2022.03 | 71.5 | 46.9 | |
| OPV2VBackbone=PointPillar, Voxel resolution=0.4 m, Compression rate=32, Fusion strategy=Intermediate2022.03 | 70.9 | 48.7 | |
| No FusionBackbone=PointPillar, Voxel resolution=0.4 m2022.03 | 60.6 | 40.2 | |
| Late FusionBackbone=PointPillar, Voxel resolution=0.4 m2022.03 | 54.9 | 30.7 |