Faster-adnet for visual tracking

Tiansa Zhang, Chunlei Huo, Zhiqiang Zhou*, Bo Wang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

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.

Original languageEnglish
Pages (from-to)684-687
Number of pages4
JournalIEICE Transactions on Information and Systems
VolumeE102D
Issue number3
DOIs
Publication statusPublished - 1 Mar 2019

Keywords

  • Deep learning
  • Status re-identification
  • Visual tracking

Fingerprint

Dive into the research topics of 'Faster-adnet for visual tracking'. Together they form a unique fingerprint.

Cite this