Context-sensitive dynamic ordinal regression for intensity estimation of facial action units
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2176
- Type
- D - Journal article
- DOI
-
10.1109/TPAMI.2014.2356192
- Title of journal
- IEEE transactions on Pattern Analysis and Machine Intelligence
- Article number
- -
- First page
- 944
- Volume
- 37
- Issue
- 5
- ISSN
- 2160-9292
- Open access status
- Out of scope for open access requirements
- Month of publication
- September
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
10.1109/TPAMI.2014.2356192
- 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
-
2
- Research group(s)
-
-
- Citation count
- 65
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper proposes the world first context-sensitive approach to estimation of intensities of facial expressions. It was applied to facial gestures (smiles, frowns) and integrated in the sentiment-based recognition system of RealEyes (in EC H2020 SEWA project, won EU Innovation Radar Award 2016; https://ec.europa.eu/digital-single-market/en/news/2016-innovation-radar-prize-winners). RealEyes analyses people's reactions to adverts from videos; Pantic is Scientific Advisor to RealEyes since 2015. Led to invited talk about Automatic Emotional Intelligence and Context-Sensitive AI at World Economic Forum 2016 (https://webcasts.weforum.org/widget/1/davos2016?p=1&pi=1&hl=english&id=76123) and to a TEDx talk on this method and the SEWA project at European Commission's Digital Assembly 2016 (https://ec.europa.eu/digital-single-market/en/programme-digital-assembly-2016).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
