Matting & Compositing

Image Matting with Bayesian/Robust Method

[1] proposed a new Bayesian framework for solving the matting problem, i.e. extracting a foreground element from a background image by estimating an opacity for each pixel of the foreground element. This approach models both the foreground and background color distributions with spatially-varying mixtures of Gaussians, and assumes a fractional blending of the foreground and background colors to produce the final output. It then uses a maximum-likelihood criterion to estimate the optimal opacity, foreground and background simultaneously.

Fig. 1, Bayesian Matting Algorithm

[2] proposed a robust matting algorithm. It also samples 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 defined energy function contains a neighborhood term to enforce the smoothness of the matte.

Fig. 2, Robust matting method

Both Bayesian and Robust Matting utilize the foreground and background colors to predict the current unknown region's alpha. Two main differences are: 1. the method of pixel sampling; 2. the way of computing alpha from foreground and background.The former is a sampling approach (solve the alphamap point by point) and the latter is a combination of sampling approach and propagation approach (solve the alphamap globally). Robust matting utilizes random walk to make its alphamap more globally reasonable.

Fig. 3, Some Results

References

1. Yung-Yu Chuang, Brian Curless, David H. Salesin, and Richard Szeliski. A Bayesian Approach to Digital Matting, IEEE CVPR, 2001.

2. Wang, M. Cohen, Optimized Color Sampling for Robust Matting , CVPR, 2007.