I'm getting much slower inference time (roughly 50% slower) inference time on a pillars + center head model versus a regular pillars + anchor 3d head model. Theoretically the center head should be faster, since it is anchor free. Is there anything that I'm doing wrong?
I have tried profiling each stage, and it looked like the non-maximum suppression in the pillars + center head model is causing the bulk of the slowdown.
I'm getting much slower inference time (roughly 50% slower) inference time on a pillars + center head model versus a regular pillars + anchor 3d head model. Theoretically the center head should be faster, since it is anchor free. Is there anything that I'm doing wrong?
I have tried profiling each stage, and it looked like the non-maximum suppression in the pillars + center head model is causing the bulk of the slowdown.
I'm running both on the KITTI dataset.
This is what the configs look like
Pillars + Center Head:
Pillars + Anchor3D Head (effectively, PointPillars):