期刊
SENSORS
卷 18, 期 10, 页码 -出版社
MDPI
DOI: 10.3390/s18103337
关键词
3D object detection; convolutional neural networks; LIDAR; autonomous driving
资金
- National Natural Science Foundation of China [61571068]
- Innovative Research Projects of Colleges and Universities in Chongqing [12A19369]
LiDAR-based or RGB-D-based object detection is used in numerous applications, ranging from autonomous driving to robot vision. Voxel-based 3D convolutional networks have been used for some time to enhance the retention of information when processing point cloud LiDAR data. However, problems remain, including a slow inference speed and low orientation estimation performance. We therefore investigate an improved sparse convolution method for such networks, which significantly increases the speed of both training and inference. We also introduce a new form of angle loss regression to improve the orientation estimation performance and a new data augmentation approach that can enhance the convergence speed and performance. The proposed network produces state-of-the-art results on the KITTI 3D object detection benchmarks while maintaining a fast inference speed.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据