Fine-Grained Pillar Feature Encoding Via Spatio-Temporal Virtual Grid for 3D Object Detection

arXiv:2403.06433v1 Announce Type: new
Abstract: Developing high-performance, real-time architectures for LiDAR-based 3D object detectors is essential for the successful commercialization of autonomous vehicles. Pillar-based methods stand out as a practical choice for onboard deployment due to their computational efficiency. However, despite their efficiency, these methods can sometimes underperform compared to alternative point encoding techniques such as Voxel-encoding or PointNet++. We argue that current pillar-based methods have not sufficiently captured the fine-grained distributions of LiDAR points within each pillar structure. Consequently, …

3d object 3d object detection abstract architectures arxiv autonomous autonomous vehicles computational cs.ai cs.cv deployment detection efficiency encoding feature fine-grained grid however lidar object performance practical real-time temporal type vehicles via virtual