LiDAR R-CNN: An Efficient and Universal 3D Object Detector
arXiv:2103.15297
Abstract
LiDAR-based 3D detection in point cloud is essential in the perception system of autonomous driving. In this paper, we present LiDAR R-CNN, a second stage detector that can generally improve any existing 3D detector. To fulfill the real-time and high precision requirement in practice, we resort to point-based approach other than the popular voxel-based approach. However, we find an overlooked issue in previous work: Naively applying point-based methods like PointNet could make the learned features ignore the size of proposals. To this end, we analyze this problem in detail and propose several methods to remedy it, which bring significant performance improvement. Comprehensive experimental results on real-world datasets like Waymo Open Dataset (WOD) and KITTI dataset with various popular detectors demonstrate the universality and superiority of our LiDAR R-CNN. In particular, based on one variant of PointPillars, our method could achieve new state-of-the-art results with minor cost. Codes will be released at https://github.com/tusimple/LiDAR_RCNN .
CVPR 2021 camera-ready
References in corpus (9)
- Vehicle Detection from 3D Lidar Using Fully Convolutional Network
- Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation
- AFDet: Anchor Free One Stage 3D Object Detection
- CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
- Voxel-FPN: multi-scale voxel feature aggregation in 3D object detection from point clouds
- Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
- Pillar-based Object Detection for Autonomous Driving
- SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds
- DMLO: Deep Matching LiDAR Odometry