3D fully convolutional network for vehicle detection in point cloud

• Published in: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) pp. 1513 - 1518
• Main Author: Bo Li
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-09-2017
• Subjects: Laser radar; Machine learning; Object detection; Three-dimensional displays; Two dimensional displays; Vehicle detection
• ISSN: 2153-0866
• DOI: 10.1109/IROS.2017.8205955

2D fully convolutional network has been recently successfully applied to the object detection problem on images. In this paper, we extend the fully convolutional network based detection techniques to 3D and apply it to point cloud data. The proposed approach is verified on the task of vehicle detection from lidar point cloud for autonomous driving. Experiments on the KITTI dataset shows significant performance improvement over the previous point cloud based detection approaches.