Abstract

Image matting is the problem of determining for each pixel in an image whether it is foreground, background, or the mixing parameter, ”alpha”, for those pixels that are a mixture of foreground and background. Matting is inherently an ill-posed problem. Previous matting approaches either use naive color sampling methods to estimate foreground and background colors for unknown pixels, or use propagation-based methods to avoid color sampling under weak assumptions about image statistics. We argue that neither method itself is enough to generate good results for complex natural images. We analyze the weaknesses of previous matting approaches, and propose a new robust matting algorithm. In our approach we also sample foreground and background colors for unknown pixels, but more importantly, analyze the confidence of these samples. Only high confidence samples are chosen to contribute to the matting energy function which is minimized by a Random Walk. The energy function we define also contains a neighborhood term to enforce the smoothness of the matte. To validate the approach, we present an extensive and quantitative comparison between our algorithm and a number of previous approaches in hopes of providing a benchmark for future matting research. 1.

Citations

372	GrabCut: Interactive foreground extraction using iterated graph cuts - Rother, Kolmogorov, et al. - 2004
138	A bayesian approach to digital matting - Chuang, Curless, et al. - 2001
116	Random walks for image segmentation - Grady - 2006
98	Blue screen matting - Smith, Blinn - 1996
76	Alpha estimation in natural images - Ruzon, Tomasi
63	M.F.: An iterative optimization approach for unified image segmentation and matting - Wang, Cohen
61	A closed form solution to natural image matting - Levin, Lischinski, et al. - 2007
58	Poisson matting - Sun, Jia, et al.
48	Knockout user guide - CORPORATION - 2002
27	Random Walks for Interactive Alpha-Matting - Grady, Schiwietz, et al. - 2005
9	Soft edge chroma-key generation based upon hexoctahedral color space - Mishima - 1993
7	Flash matting - Sun, Li, et al. - 2006
4	Easy matting - Guan, Chen, et al. - 2006
2	Multilabel randomwalker image segmentation using prior models - Grady - 2005