V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction

Benchmark Model Rank Results
3d-object-detection-on-opv2vV2VNet (PointPillar backbone)#1AP@0.7@Default: 0.822AP@0.7@CulverCity: 0.734
3d-object-detection-on-v2x-simV2VNet#4mAP: 21.4mATE: 0.768mASE: 0.255mAOE: 0.349
3d-object-detection-on-v2xsetV2VNet#3AP0.5 (Perfect): 0.845AP0.7 (Perfect): 0.677AP0.5 (Noisy): 0.791