IMAGE MATTING USING SVM AND NEIGHBORING INFORMATION

Tadaaki Hosaka, Takumi Kobayashi, Nobuyuki Otsu

Abstract

Image matting is a technique for extracting a foreground object in a static image by estimating the opacity at each pixel in the foreground image layer. This problem has recently been studied in the framework of optimizing a cost function. The common drawback of previous approaches is the decrease in performance when the foreground and background contain similar colors. To solve this problem, we propose a cost function considering not only a single pixel but also its neighboring pixels, and utilizing the SVM classifier to enhance the discrimination between the foreground and background. Optimization of the cost function can be achieved by belief propagation. Experimental results show favorable matting performance for many images.

Download


Paper Citation


in Harvard Style

Hosaka T., Kobayashi T. and Otsu N. (2007). IMAGE MATTING USING SVM AND NEIGHBORING INFORMATION . In Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, ISBN 978-972-8865-73-3, pages 344-349. DOI: 10.5220/0002060903440349


in Bibtex Style

@conference{visapp07,
author={Tadaaki Hosaka and Takumi Kobayashi and Nobuyuki Otsu},
title={IMAGE MATTING USING SVM AND NEIGHBORING INFORMATION},
booktitle={Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,},
year={2007},
pages={344-349},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002060903440349},
isbn={978-972-8865-73-3},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP,
TI - IMAGE MATTING USING SVM AND NEIGHBORING INFORMATION
SN - 978-972-8865-73-3
AU - Hosaka T.
AU - Kobayashi T.
AU - Otsu N.
PY - 2007
SP - 344
EP - 349
DO - 10.5220/0002060903440349