Closed StiphyJay closed 1 year ago
All the experiments are evaluated on the same workstation (an NVIDIA A100 GPU with AMD EPYC 7513 CPU). We just follow the implementation of OpenPCDet centerpoint-pillar. Its inference time contains hard-voxelization(cpu) + PillarVFE + backbone2D + centerhead + nms.
Congratulations for accepted in CVPR! I have a problem with Fig. 1. In your paper, your method is evaluated on an NVIDIA A100 GPU. But for other methods, you also evaluated them on A100 GPU? Especially for PointPillars, and CenterPoint-Pillar. Did you also evaluate their speed at the same A100 GPU device? The CenterPoint-Pillar's speed is only 30FPS, which seems strange. A100 has a higher computing power.