Beyond triplet loss : a deep quadruplet network for person re-identification
- Submitting institution
-
University of Dundee
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 28402995
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR.2017.145
- Title of conference / published proceedings
- Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition
- First page
- 1320
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- November
- Year of publication
- 2017
- 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
-
3
- Research group(s)
-
-
- Citation count
- 188
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- A quadruplet loss is proposed addressing weaknesses of the widely used triplet loss. The novel quadruplet deep network design has inspired development of a wider set of techniques in computer vision and beyond, including systems for traffic sign recognition at MIT (Kim et al., AAAI, 2018), vehicle re-identification (Hou et al., IEEE Trans. Vehicular Technology, 2019), clothing retrieval (Miao et al., IEEE Access, 2020), remote sensing (Zhang et al. Remote Sensing, 2020), and Tencent’s CosFace algorithm for face recognition (Wang et al., CVPR, 2018).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -