Gaze-Informed Egocentric Action Recognition for Memory Aid Systems
- Submitting institution
-
Manchester Metropolitan University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2394
- Type
- D - Journal article
- DOI
-
10.1109/access.2018.2808486
- Title of journal
- IEEE Access
- Article number
- -
- First page
- 12894
- Volume
- 6
- Issue
- -
- ISSN
- 2169-3536
- Open access status
- Compliant
- Month of publication
- March
- Year of publication
- 2018
- URL
-
https://e-space.mmu.ac.uk/627183/
- 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
-
4
- Research group(s)
-
A - Data Science
- Citation count
- 8
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Egocentric action recognition in video plays a vital role in many applications including pervasive healthcare. This work is significant as it is the first attempt at using real-time gaze information to actively track people’s action in the loop for adaptive recognition of the regions of interest (ROI). It overcomes the limitations of existing approaches that require a full video frame or pre-defined ROI. An empirical evaluation reveals that the approach has significant benefits as it is both highly effective and accurate underlining the potential of the memory support tool in health care.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -