DMCA
Optimized color sampling for robust matting (2007)
Cached
Download Links
- [juew.org]
- [research.microsoft.com]
- [www.research.microsoft.com]
- [grail.cs.washington.edu]
- [research.microsoft.com]
- [research.microsoft.com]
- [www.juew.org]
- DBLP
Other Repositories/Bibliography
| Venue: | In In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition |
| Citations: | 75 - 5 self |
BibTeX
@INPROCEEDINGS{Wang07optimizedcolor,
author = {Jue Wang and Michael F. Cohen},
title = {Optimized color sampling for robust matting},
booktitle = {In In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition},
year = {2007}
}
Share
 |
 |
 |
 |
||
OpenURL
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 inher-ently an ill-posed problem. Previous matting approaches either use naive color sampling methods to estimate fore-ground and background colors for unknown pixels, or use propagation-based methods to avoid color sampling under weak assumptions about image statistics. We argue that nei-ther method itself is enough to generate good results for complex natural images. We analyze the weaknesses of previous matting ap-proaches, 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 sam-ples are chosen to contribute to the matting energy function which is minimized by a Random Walk. The energy func-tion we define also contains a neighborhood term to en-force the smoothness of the matte. To validate the approach, we present an extensive and quantitative comparison be-tween our algorithm and a number of previous approaches in hopes of providing a benchmark for future matting re-search. 1.
Keyphrases
robust matting background color unknown pixel propagation-based method mixing parameter good result high confidence sam-ples random walk image matting energy func-tion new robust matting algorithm previous approach energy function naive color complex natural image quantitative comparison neighborhood term image statistic weak assumption ill-posed problem
