Optimized Color Sampling for Robust Matting Export

@inproceedings{citeulike:3196284, 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.}, author = {Wang, Jue and Cohen, M. F.}, booktitle = {Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on}, citeulike-article-id = {3196284}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/CVPR.2007.383006}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4270031}, doi = {10.1109/CVPR.2007.383006}, journal = {Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on}, keywords = {alpha, matting, optimal, robust, sampling, segmentation}, pages = {1--8}, posted-at = {2008-09-04 21:57:05}, priority = {5}, title = {Optimized Color Sampling for Robust Matting}, url = {http://dx.doi.org/10.1109/CVPR.2007.383006}, year = {2007} }

TY - CONF ID - citeulike:3196284 L3 - citeulike-article-id:3196284 N2 - 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. JF - Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on EP - 8 TI - Optimized Color Sampling for Robust Matting T2 - Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on SP - 1 KW - alpha KW - matting KW - optimal KW - robust KW - sampling KW - segmentation AU - Wang, Jue AU - Cohen, MF PY - 2007/// UR - http://dx.doi.org/10.1109/CVPR.2007.383006 ER -