PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

Benchmark Model Rank Results
3d-object-detection-on-kitti-cars-easyPV-RCNN#6AP: 90.25%
3d-object-detection-on-kitti-cars-hardPV-RCNN#7AP: 76.82%
3d-object-detection-on-kitti-cyclistsPV-RCNN#3AP: 63.71%
3d-object-detection-on-kitti-cyclists-easyPV-RCNN#3AP: 78.60%
3d-object-detection-on-kitti-cyclists-hardPV-RCNN#3AP: 57.65%
3d-object-detection-on-waymo-cyclistPV-RCNN#5APH/L2: 71.16
3d-object-detection-on-waymo-pedestrianPV-RCNN#6APH/L2: 70.16
3d-object-detection-on-waymo-vehiclePV-RCNN#4APH/L2: 73.23
birds-eye-view-object-detection-on-kittiPV-RCNN#1AP: 68.89%
birds-eye-view-object-detection-on-kitti-carsPV-RCNN#2AP: 90.65%
birds-eye-view-object-detection-on-kitti-cars-4PV-RCNN#2AP: 94.98
birds-eye-view-object-detection-on-kitti-cars-5PV-RCNN#2AP: 86.14