Inpainting of Wide-Baseline Multiple Viewpoint Video
- Submitting institution
-
The University of Surrey
- Unit of assessment
- 33 - Music, Drama, Dance, Performing Arts, Film and Screen Studies
- Output identifier
- 9007966_1
- Type
- D - Journal article
- DOI
-
10.1109/TVCG.2018.2889297
- Title of journal
- IEEE Transactions on Visualization and Computer Graphics
- Article number
- -
- First page
- 2417
- Volume
- 26
- Issue
- 7
- ISSN
- 1077-2626
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- 12 - Engineering
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Factual information about significance: This paper presents the first algorithm for wide-baseline MVV inpainting and corrects the distortion that would normally make this approach impractical. For the first time it is now possible to peek behind large object occlusions to inpaint background elements never revealed to the principal camera. Performance is state-of-the-art in terms of both accuracy and computational speed. Future generative-based inpainting systems can make use of this research to achieve superior temporal and spatial consistency, even with dynamic and cluttered scenes.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -