Reconfigurable Voxels: A New Representation for LiDAR-Based Point Clouds
This addresses a critical problem in autonomous driving by enhancing 3D object detection for sparse regions, though it appears incremental as it builds on existing voxel-based methods.
The paper tackles the challenge of detecting small and distant objects in sparse, irregular LiDAR point clouds for autonomous driving by proposing Reconfigurable Voxels, which improves detection performance on benchmarks like nuScenes, Lyft, and KITTI without significant overhead.
LiDAR is an important method for autonomous driving systems to sense the environment. The point clouds obtained by LiDAR typically exhibit sparse and irregular distribution, thus posing great challenges to the detection of 3D objects, especially those that are small and distant. To tackle this difficulty, we propose Reconfigurable Voxels, a new approach to constructing representations from 3D point clouds. Specifically, we devise a biased random walk scheme, which adaptively covers each neighborhood with a fixed number of voxels based on the local spatial distribution and produces a representation by integrating the points in the chosen neighbors. We found empirically that this approach effectively improves the stability of voxel features, especially for sparse regions. Experimental results on multiple benchmarks, including nuScenes, Lyft, and KITTI, show that this new representation can remarkably improve the detection performance for small and distant objects, without incurring noticeable overhead costs.