4 March 2015 An improved Bayesian matting method based on image statistic characteristics
Author Affiliations +
Proceedings Volume 9443, Sixth International Conference on Graphic and Image Processing (ICGIP 2014); 94431F (2015) https://doi.org/10.1117/12.2179335
Event: Sixth International Conference on Graphic and Image Processing (ICGIP 2014), 2014, Beijing, China
Abstract
Image matting is an important task in image and video editing and has been studied for more than 30 years. In this paper we propose an improved interactive matting method. Starting from a coarse user-guided trimap, we first perform a color estimation based on texture and color information and use the result to refine the original trimap. Then with the new trimap, we apply soft matting process which is improved Bayesian matting with smoothness constraints. Experimental results on natural image show that this method is useful, especially for the images have similar texture feature in the background or the images which is hard to give a precise trimap.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Wei Sun, Wei Sun, Siwei Luo, Siwei Luo, Lina Wu, Lina Wu, } "An improved Bayesian matting method based on image statistic characteristics", Proc. SPIE 9443, Sixth International Conference on Graphic and Image Processing (ICGIP 2014), 94431F (4 March 2015); doi: 10.1117/12.2179335; https://doi.org/10.1117/12.2179335
PROCEEDINGS
5 PAGES


SHARE
RELATED CONTENT

Learning historical heritage with a serious game a...
Proceedings of SPIE (March 04 2015)
An interactive tool for gamut masking
Proceedings of SPIE (February 24 2014)
The ontology model of frontCRM framework
Proceedings of SPIE (March 20 2013)
Depth keying
Proceedings of SPIE (May 30 2003)
Maximizing inpainting efficiency without sacrificing quality
Proceedings of SPIE (January 18 2010)
Temporal coherency for video tone mapping
Proceedings of SPIE (October 15 2012)

Back to Top