Abstract
We present a technique for colour matching between stereo pairs of images. We use a pre-computed disparity map to find corresponding areas and achieve high quality results for localised colour grading. We avoid direct pixel-to-pixel matches and instead use a gridding system in order to constrain the area of focus and allow for a more subtle manipulation of colour. Our method maintains existing noise without introducing more, allowing it to be used in movie post-production without affecting other artistic decisions. We bring together elements from image correspondence algorithms and colour transfer techniques in order to manipulate localised areas to have a very close match.
Original language | English |
---|---|
Title of host publication | Proceedings of the 13th European Conference on Visual Media Production, CVMP 2016 |
Publisher | Association for Computing Machinery |
ISBN (Print) | 9781450347440 |
DOIs | |
Publication status | Published - 12 Dec 2016 |
Event | 13th European Conference on Visual Media Production, CVMP 2016 - London, UK United Kingdom Duration: 12 Dec 2016 → 13 Dec 2016 |
Conference
Conference | 13th European Conference on Visual Media Production, CVMP 2016 |
---|---|
Country/Territory | UK United Kingdom |
City | London |
Period | 12/12/16 → 13/12/16 |
Funding
With many thanks to the project supervisors for their support and advice, the anonymous reviewers for their helpful comments and to the Centre for Digital Entertainment, Double Negative Ltd. and EPSRC for supporting this research. The Centre for Digital Entertainment is a national Doctoral Training Centre and we gratefully acknowledge substantial funding from the EPSRC Digital Economy programme.
Keywords
- Colour correction
- Image processing
- Stereo
ASJC Scopus subject areas
- Software
- Human-Computer Interaction
- Computer Vision and Pattern Recognition
- Computer Networks and Communications