Faster-ADNet for Visual Tracking

Tiansa ZHANG  Chunlei HUO  Zhiqiang ZHOU  Bo WANG  

IEICE TRANSACTIONS on Information and Systems   Vol.E102-D   No.3   pp.684-687
Publication Date: 2019/03/01
Publicized: 2018/12/12
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2018EDL8214
Type of Manuscript: LETTER
Category: Image Recognition, Computer Vision
visual tracking,  deep learning,  status re-identification,  

Full Text: PDF>>
Buy this Article

By taking advantages of deep learning and reinforcement learning, ADNet (Action Decision Network) outperforms other approaches. However, its speed and performance are still limited by factors such as unreliable confidence score estimation and redundant historical actions. To address the above limitations, a faster and more accurate approach named Faster-ADNet is proposed in this paper. By optimizing the tracking process via a status re-identification network, the proposed approach is more efficient and 6 times faster than ADNet. At the same time, the accuracy and stability are enhanced by historical actions removal. Experiments demonstrate the advantages of Faster-ADNet.