Geometrical-based lip-reading using template probabilistic multi-dimension dynamic time warping
- Submitting institution
-
Loughborough University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1936
- Type
- D - Journal article
- DOI
-
10.1016/j.jvcir.2015.04.013
- Title of journal
- Journal of Visual Communication and Image Representation
- Article number
- -
- First page
- 219
- Volume
- 30
- Issue
- -
- ISSN
- 1047-3203
- Open access status
- Out of scope for open access requirements
- Month of publication
- May
- Year of publication
- 2015
- 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
-
1
- Research group(s)
-
-
- Citation count
- 10
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper describes the first results from a new speech recognition corpus developed at Loughborough University with support from the Malaysian Government. The high-resolution images, coupled with novel combinations of audio and visual information, enabled a substantial improvement in state-of-the-art audio-visual recognition performance, even compared with video-only recognition in good lighting conditions (a research first). The new corpus has since been used by a number of researchers in University Malaysia Pahang, leading to an additional ten collaborative publications (see https://scholar.google.co.uk/citations?hl=en&user=KYnHdrYAAAAJ&view_op=list_works&sortby=pubdate).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -