Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing
- Submitting institution
-
Aberystwyth University / Prifysgol Aberystwyth
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 38986647
- Type
- D - Journal article
- DOI
-
10.1109/TCYB.2016.2608906
- Title of journal
- IEEE Transactions on Cybernetics
- Article number
- 7579645
- First page
- 4342
- Volume
- 12
- Issue
- 47
- ISSN
- 2168-2267
- Open access status
- Technical exception
- Month of publication
- September
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
https://ieeexplore.ieee.org/ielx7/6221036/8101048/7579645/tcyb-lin-2608906-mm.zip?tp=&arnumber=7579645
- 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
- 76
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Presents an underpinning approach for modelling instances across different modalities in a binary space. The paper shows for the first time that it is theoretically possible to generate one unified hash code for all observed views of any instance, enabling fast cross-view data retrieval. Published in TCYB, a top-rated outlet across all CS&I areas (acceptance rate <10%). The method has become a standard reference for research on cross-modal data retrieval (Kittler et al. 2018). Various new methods (Mandal et al. 2018, IIS; Hu et al. 2020, I2R; Zhang et al. 2020, PekingU) are based on this work.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
