Pencil drawing video rendering using convolutional networks
- Submitting institution
-
Liverpool John Moores University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1011
- Type
- D - Journal article
- DOI
-
10.1111/cgf.13819
- Title of journal
- Computer Graphics Forum: the international journal of the Eurographics Association
- Article number
- -
- First page
- 91
- Volume
- 38
- Issue
- 7
- ISSN
- 0167-7055
- Open access status
- Compliant
- Month of publication
- November
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
https://github.com/Kanata-Bifang/CNN-for-video-transfer
- 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
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Pencil drawing video rendering is addressed for the first time. Taking advantage of Convolutional Neural Networks (CNNs), this work presents a new feed-forward CNN to resolve the problems of temporal inconsistency and shower-door effect caused by traditional spatial algorithms. The developed CNN can also be applied to other art-style video rendering. The work is a result of an international research collaboration with East China Normal University and Hangzhou Dianzi University in China as well as the University of Yamanashi in Japan. Its open source code has been made publically available (https://github.com/Kanata-Bifang/CNN-for-video-transfer).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
