Learning Adaptive Discriminative Correlation Filters via Temporal Consistency Preserving Spatial Feature Selection for Robust Visual Object Tracking
- Submitting institution
-
The University of Surrey
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 9018801_2
- Type
- D - Journal article
- DOI
-
10.1109/TIP.2019.2919201
- Title of journal
- IEEE Transactions on Image Processing
- Article number
- -
- First page
- 5596
- Volume
- 28
- Issue
- 11
- ISSN
- 1057-7149
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 50
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This is the winning algorithm of the Visual Object Tracking (VOT2018) competition on the public dataset, among 72 algorithms. VOT is the most famous competition in the visual tracking community, which is held annually. In this work, we propose a spatial feature selection method for discriminative-correlation-filter-based visual object tracking. The proposed method improves the tracking performance significantly as compared with the state-of-the-art approaches. The idea of spatial feature selection proposed in this work has been followed by many researchers in the visual object tracking community.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -