Deep neural network augmentation: generating faces for affect analysis
- Submitting institution
-
University of Greenwich
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 29423
- Type
- D - Journal article
- DOI
-
10.1007/s11263-020-01304-3
- Title of journal
- International Journal of Computer Vision
- Article number
- -
- First page
- 1455
- Volume
- 128
- Issue
- 5
- ISSN
- 0920-5691
- Open access status
- Compliant
- Month of publication
- February
- Year of publication
- 2020
- 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
-
4
- Research group(s)
-
-
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Generating high-quality facial images/image-sequences with dimensional or categorical affect is of great significance for data augmentation in DL-based affect prediction. Our approach outperformed the state-of-the-art (based on GANs). The facial affect generation part was also presented at Human Behavior Understanding Workshop, held in conjunction with ECCV 2018. The data augmentation part attracted the interest of Deep Render Ltd company; we collaborated for applying our approach to generate images and use them for data augmentation in training models for image compression.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -