3DSSD: Point-based 3D Single Stage Object Detector

TL;DR

This paper introduces 3DSSD, a lightweight point-based 3D object detector that balances accuracy and efficiency by removing complex upsampling and refinement stages, and employs a novel fusion sampling strategy for better detection.

Contribution

The paper presents a novel single-stage, anchor-free point-based 3D object detection framework that improves speed and accuracy over existing methods.

Findings

Outperforms voxel-based single-stage methods significantly.

Achieves comparable results to two-stage point-based methods.

Runs at over 25 FPS, doubling the speed of previous point-based detectors.

Abstract

Currently, there have been many kinds of voxel-based 3D single stage detectors, while point-based single stage methods are still underexplored. In this paper, we first present a lightweight and effective point-based 3D single stage object detector, named 3DSSD, achieving a good balance between accuracy and efficiency. In this paradigm, all upsampling layers and refinement stage, which are indispensable in all existing point-based methods, are abandoned to reduce the large computation cost. We novelly propose a fusion sampling strategy in downsampling process to make detection on less representative points feasible. A delicate box prediction network including a candidate generation layer, an anchor-free regression head with a 3D center-ness assignment strategy is designed to meet with our demand of accuracy and speed. Our paradigm is an elegant single stage anchor-free framework, showing great superiority to other existing methods. We evaluate 3DSSD on widely used KITTI dataset and more challenging nuScenes dataset. Our method outperforms all state-of-the-art voxel-based single stage methods by a large margin, and has comparable performance to two stage point-based methods as well, with inference speed more than 25 FPS, 2x faster than former state-of-the-art point-based methods.