A time flexible kernel framework for video-based activity recognition
- Submitting institution
-
Kingston University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-35-1367
- Type
- D - Journal article
- DOI
-
10.1016/j.imavis.2015.12.006
- Title of journal
- Image and Vision Computing
- Article number
- -
- First page
- 26
- Volume
- 48-49
- Issue
- -
- ISSN
- 0262-8856
- Open access status
- Out of scope for open access requirements
- Month of publication
- -
- 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
-
-
- Research group(s)
-
-
- Citation count
- 8
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- A major challenge in activity recognition is time encoding. Indeed, time information may be lost through video encoding (e.g. when bags of features are used) or when a classification method requires a fixed-size vector per activity as input, which makes comparison of activities of varied duration difficult. This paper is significant because the proposed Time-Flexible Kernel framework improves the performance of activity recognition and widen the applicability of such systems.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -