4.6 Article

SECOND: Sparsely Embedded Convolutional Detection

Journal

SENSORS
Volume 18, Issue 10, Pages -

Publisher

MDPI
DOI: 10.3390/s18103337

Keywords

3D object detection; convolutional neural networks; LIDAR; autonomous driving

Funding

  1. National Natural Science Foundation of China [61571068]
  2. Innovative Research Projects of Colleges and Universities in Chongqing [12A19369]

Ask authors/readers for more resources

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.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available