Separating a foreground object from the background in a static image involves determining both full and partial pixel coverages, also known as extracting a matte. Previous approaches require the input image to be presegmented into three regions: foreground, background and unknown, which are called a trimap. Partial opacity values are then computed only for pixels inside the unknown region. This presegmentation based approach fails for images with large portions of semitransparent foreground where the trimap is difficult to create even manually. In this paper, we combine the segmentation and matting problem together and propose a unified optimization approach based on belief propagation. We iteratively estimate the opacity value for every pixel in the image, based on a small sample of foreground and background pixels marked by the user. Experimental results show that compared with previous approaches, our method is more efficient to extract high quality mattes for foregrounds with significant semitransparent regions.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    An iterative optimization approach for unified image segmentation and matting


    Contributors:
    Wang, J. (author) / Cohen, M.F. (author)


    Publication date :

    2005-01-01


    Size :

    1232868 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    An Iterative Optimization Approach for Unified Image Segmentation and Matting

    Wang, J. / Cohen, M. / IEEE | British Library Conference Proceedings | 2005



    Accurate motion layer segmentation and matting

    Xiao, J. / Shah, M. | IEEE | 2005


    Automatic video matting based on hybrid video object segmentation and closed-form matting

    Hu, W.-C. / Hsu, J.-F. / Huang, D.-Y. | British Library Online Contents | 2013


    Image Matting for Automatic Target Recognition

    Hyun-Woong Cho / Young-Rae Cho / Woo-Jin Song et al. | IEEE | 2017