Learnt quasi-transitive similarity for retrieval from large collections of faces
- Submitting institution
-
University of St Andrews
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 251887714
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR.2016.528
- Title of conference / published proceedings
- Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- First page
- 4883
- Volume
- -
- Issue
- -
- ISSN
- 1063-6919
- Open access status
- Out of scope for open access requirements
- Month of publication
- June
- Year of publication
- 2016
- 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)
-
A - Artificial Intelligence
- Citation count
- 6
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- In this paradigmatically most novel paper, it is shown that what was generally seen as the central challenge in face recognition, and indeed classification in general -- that is inter-personal (or inter-class) similarity -- can actually be beneficial in the retrieval setting. This highly disruptive finding is first demonstrated on purely theorical grounds, and then empirically, and the first method that formalizes the idea introduced.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
