Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression
- Submitting institution
-
University of Nottingham, The
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1330524
- Type
- E - Conference contribution
- DOI
-
10.1109/ICCV.2017.117
- Title of conference / published proceedings
- 2017 IEEE International Conference on Computer Vision
- First page
- 1031
- Volume
- 2017-October
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- December
- Year of publication
- 2017
- 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
-
3
- Research group(s)
-
-
- Citation count
- 84
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper presents a novel approach to 3D face reconstruction resilient to occlusion, poor lighting and large pose, using only a single facial image. Alternative approaches require construction and alignment of morphable models, careful initialisation, or multiple images. Published at a high-ranking conference (acceptance rate 26%), the method has been widely used, generating >1.5 million 3D faces through the online demo alone. 3D prints generated by the method have been exhibited in art galleries. It has been integrated into Wolfram Mathematica Neural Net Repository and received broad media coverage, including TV interviews, articles in Mail Online, Mashable and many others.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
