22 July 2021 Structure aware 3D single object tracking of point cloud
Xiaoyu Zhou, Ling Wang, Zhian Yuan, Ke Xu, Yanxin Ma
Author Affiliations +
Abstract
Existing 3D single object trackers (SOTs) of a point cloud all apply downscaling when extracting features from points. This operation leads to a loss of spatial and structural information, degrading tracking performance of sparsely distributed and small-scale objects. To address this problem, a structure aware SOT of a point cloud is proposed. Specifically, the backbone network is combined with the auxiliary network to learn point-wise representations. During the training stage, the subsidiary network is used to perform additional tasks and supervisions, which guides the backbone network to extract discriminative structural features. During the inference stage, this network part is detached to meet a real-time requirement as well as to ensure the tracking accuracy. In addition, the impacts of the quantity setting of the input point cloud and re-initiation strategy are discussed; these are significant to the performance but have been ignored by former works. The experimental results show that the proposed method has a distinct improvement even if the tracked object is sparse and small scale.
© 2021 SPIE and IS&T 1017-9909/2021/$28.00 © 2021 SPIE and IS&T
Xiaoyu Zhou, Ling Wang, Zhian Yuan, Ke Xu, and Yanxin Ma "Structure aware 3D single object tracking of point cloud," Journal of Electronic Imaging 30(4), 043010 (22 July 2021). https://doi.org/10.1117/1.JEI.30.4.043010
Received: 1 April 2021; Accepted: 30 June 2021; Published: 22 July 2021
Lens.org Logo
CITATIONS
Cited by 4 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Clouds

Image segmentation

Feature extraction

LIDAR

Sensors

Visualization

3D acquisition

Back to Top