Spatial information enhancement network for 3D object detection from point cloud. (August 2022)

• Record Type: Journal Article
• Authors: Li, Ziyu; Yao, Yuncong; Quan, Zhibin; Xie, Jin; Yang, Wankou
• Is Part Of: Pattern recognition. Volume 128(2022)
• Publication Date: 2022-08
• Subjects: 3D object detection -- Autonomous vehicles -- Point cloud -- LiDAR sensor -- 3D shape completion; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.108684
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: To address the density imbalanced problem in point clouds, we propose a novel spatial information enhancement module (SIE) to predict the dense shapes of point sets in candidate boxes, and learn the structure information to improve the ability of feature representation. We present a hybrid-paradigm region proposal network (HP-RPN) for more effective multi-scale feature extraction and high-recall proposal generation. With the structure information as guidance, our elaborately designed SIENet achieves the state-of-the-art performance of 3D object detection on the KITTI benchmark. The encouraging experimental results also demonstrate the outstanding improvement in far-range object detection. Abstract: LiDAR-based 3D object detection pushes forward an immense influence on autonomous vehicles. Due to the limitation of the intrinsic properties of LiDAR, fewer points are collected at the objects farther away from the sensor. This imbalanced density of point clouds degrades the detection accuracy but is generally neglected by previous works. To address the challenge, we propose a novel two-stage 3D object detection framework, named SIENet. Specifically, we design the Spatial Information Enhancement (SIE) module to predict the spatial shapes of the foreground points within proposals, and extract the structure information to learn the representative features for further box refinement. The predicted spatial shapes are complete and dense point sets, thus the extracted structure information contains more semantic representation. Besides, we design the Hybrid-Paradigm Region Proposal Network (HP-RPN) which includes multiple branches to learn discriminate features and generate accurate proposals for the SIE module. Extensive experiments on the KITTI 3D object detection benchmark show that our elaborately designed SIENet outperforms the state-of-the-art methods by a large margin. Codes will be publicly available at https://github.com/Liz66666/SIENet .