Learning Perceptual Aesthetics of 3D Shapes from Multiple Views
- Submitting institution
-
Liverpool Hope University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- KD24C
- Type
- D - Journal article
- DOI
-
10.1109/MCG.2020.3026137
- Title of journal
- IEEE Computer Graphics and Applications
- Article number
- -
- First page
- 1-1
- Volume
- Early Access
- Issue
- -
- ISSN
- 1558-1756
- Open access status
- Compliant
- Month of publication
- September
- 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
- Yes
- Number of additional authors
-
1
- Research group(s)
-
S - Spatial Computing and Robotics (SC&R)
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Selected as a feature article in the journal: IEEE Computer Graphics and Applications.
In this paper, we study the perceptual aesthetics of 3D shapes. We are the first to develop a measure of 3D shape aesthetics using a multi-view deep convolution ranking network and human aesthetics preference data. We also release to the research community a large amount of human aesthetics preference data on 3D shape pairs and build tools to demonstrate the important applications in data organisation, search, and scene composition.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
