Chuantong Hou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Chuantong Hou has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Boyang Li
College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China
1 shared publication - Miao Li
College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China
1 shared publication - Nuo Chen
College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China
1 shared publication - Wei An
National University of Defense Technology, China
1 shared publication
Latest Publications
- Open Access
- 25 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Infrared moving small multi-targets trajectory detection based on point cloud instance segmentation
In this paper, a point cloud instance segmentation method is proposed to achieve infrared moving small multi-targets trajectory detection. Different from previous trajectory detection methods that used manually designed features to achieve frame-by-frame correlation, this paper first models 100 frames of candidate point images in 2D space along the time dimension as a candidate point cloud in 3D space. In this way, we can transform the problem of frame-by-frame correlation of candidate points in the 2D space as the problem of segmenting the trajectory instances with specific linear topological structures in the 3D point cloud space. To this end, we propose a simple but effective framework Point Trajectory Detection Network (namely, PTDN). Moreover, we also construct a 3D infrared moving small multi-targets candidate point cloud dataset with complex moving trajectory, varied number of small target and additional noisy trajectory. Experimental results can verify the superiority of our PTDN compared with the traditional 2D trajectory detection algorithms and 3D instance segmentation algorithms.
- 25 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessInfrared moving small multi-targets trajectory detection based on point cloud instance segmentation
In this paper, a point cloud instance segmentation method is proposed to achieve infrared moving small multi-targets trajectory detection. Different from previous trajectory detection methods that used manually designed features to achieve frame-by-frame correlation, this paper first models 100 frames of candidate point images in 2D space along the time dimension as a candidate point cloud in 3D space. In this way, we can transform the problem of frame-by-frame correlation of candidate points in the 2D space as the problem of segmenting the trajectory instances with specific linear topological structures in the 3D point cloud space. To this end, we propose a simple but effective framework Point Trajectory Detection Network (namely, PTDN). Moreover, we also construct a 3D infrared moving small multi-targets candidate point cloud dataset with complex moving trajectory, varied number of small target and additional noisy trajectory. Experimental results can verify the superiority of our PTDN compared with the traditional 2D trajectory detection algorithms and 3D instance segmentation algorithms.