Harmonious Attention Network for Person Re-Identification
- Submitting institution
-
Queen Mary University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 516
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR.2018.00243
- Title of conference / published proceedings
- Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
- First page
- 2285
- Volume
- -
- Issue
- -
- ISSN
- 1063-6919
- Open access status
- Not compliant
- Month of publication
- June
- Year of publication
- 2018
- 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
-
0
- Research group(s)
-
-
- Citation count
- 255
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Funded by Royal Society Newton Advanced Fellowship, Innovate UK Industrial Challenge Project on Developing and Commercialising Intelligent Video Analytics Solutions for Public Safety, and Vision Semantics. A fundamentally new design of deep network architecture for person Re-ID. Ranked No. 1 on performance at the largest Re-ID public benchmark Market-1501 (comparing 42 state-of-the-art models since 2016). An extension of an early work IJCAI'17. Keynotes at BMVC'18, ECCV'18. Commercial trials outperform all other models (commercial and academic) on customers' own data in different countries without model retraining on local data. Built into Vision Semantics and AccuVision products.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -