Interactive Image Segmentation via Pairwise Likelihood Learning

• DOI: 10.24963/ijcai.2017/412
• Publication Date: 2017-07-28T05:14:07Z
• Authors: Tao Wang; Quansen Sun; Qi Ge; Zexuan Ji; Qiang Chen; Guiyu Xia
• Keywords: Unary operation; Robustness; Smoothing

This paper presents an interactive image segmentation approach where the problem is formulated as a probabilistic estimation manner. Instead of measuring distances between unseeded pixels and seeded pixels, we measure similarities pixel pairs seed to improve robustness seeds. The unary prior probability each belonging foreground F background B can be effectively estimated based on with label (F, F),(F, B),(B, F) (B, B). Then likelihood learning framework proposed fuse region boundary information by imposing smoothing constraint potentials. Experiments challenging data sets demonstrate that method obtain better performance than state-of-the-art methods.