Unsupervised Deep Video Hashing via Balanced Code for Large-Scale Video Retrieval
- Submitting institution
-
Aberystwyth University / Prifysgol Aberystwyth
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 38748763
- Type
- D - Journal article
- DOI
-
10.1109/TIP.2018.2882155
- Title of journal
- IEEE Transactions on Image Processing
- Article number
- -
- First page
- 1993
- Volume
- 28
- Issue
- 4
- ISSN
- 1057-7149
- Open access status
- Technical exception
- Month of publication
- November
- 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
-
6
- Research group(s)
-
-
- Citation count
- 45
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Archiving version that substantially extends the paper presented at top AI conference IJCAI-2017 (26% acceptance rate, not returned); IEEE T-IP is the foremost outlet in image processing. Conducted in collaboration with Tsinghua University, China, and funded by Royal Society Newton mobility grant (IE150997), this work makes the problem of binarizing deep learned video features more tractable. We demonstrate, for the first time, that feature learning and hashing code learning can be jointly optimized in a uniform framework, capable of accomplishing the task of fast nearest neighbour search, given a large-scale video dataset.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -