Rethinking Voxelization and Classification for 3D Object Detection
Abstract
The main challenge in 3D object detection from LiDAR point clouds is achieving real-time performance without affecting the reliability of the network. In other words, the detecting network must be confident enough about its predictions. In this paper, we present a solution to improve network inference speed and precision at the same time by implementing a fast dynamic voxelizer that works on fast pillar-based models in the same way a voxelizer works on slow voxel-based models. In addition, we propose a lightweight detection sub-head model for classifying predicted objects and filter out false detected objects that significantly improves model precision in a negligible time and computing cost. The developed code is publicly available at: https://github.com/YoushaaMurhij/RVCDet.
Cite
@article{arxiv.2301.04058,
title = {Rethinking Voxelization and Classification for 3D Object Detection},
author = {Youshaa Murhij and Alexander Golodkov and Dmitry Yudin},
journal= {arXiv preprint arXiv:2301.04058},
year = {2023}
}
Comments
Accepted in ICONIP 2022. arXiv admin note: text overlap with arXiv:1902.06326 by other authors