Learning based alpha matting using support vector regression
Authors:
- 1 September 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15224880,p. 2109-2112
Abstract
Alpha matting refers to the problem of estimating the opacity mask of the foreground in an image. Many recent algorithms solve it with color samples or some local assumptions, causing artifacts when they fail to collect appropriate samples or the assumptions do not hold. In this paper, we treat alpha matting as a supervised learning problem and propose a new matting approach. Given the input image and a trimap (labeling some foreground/background pixels), we segment the unlabeled region into pieces and learn the relations between pixel features and alpha values for these pieces. We use support vector regression (SVR) in the learning process. To obtain better learning results, we design a training samples selection method and use adaptive parameters for SVR. Qualitative and quantitative evaluations on a matting benchmark show that our approach outperforms many recent algorithms in terms of accuracy.Keywords
This publication has 12 references indexed in Scilit:
- A global sampling method for alpha matting2011
- LIBSVMACM Transactions on Intelligent Systems and Technology, 2011
- Fast matting using large kernel matting Laplacian matrices2010
- A spatially varying PSF-based prior for alpha matting2010
- Image matting based on local color discrimination by SVMPattern Recognition Letters, 2009
- Improving Color Modeling for Alpha Matting2008
- Optimized Color Sampling for Robust Matting2007
- A Closed Form Solution to Natural Image Matting2006
- A tutorial on support vector regressionStatistics and Computing, 2004
- Practical selection of SVM parameters and noise estimation for SVM regressionNeural Networks, 2004