Error-tolerant interactive image segmentation using dynamic and iterated graph-cuts
Ozan Şener, Kemal Uǧur, A. Aydın Alatan · 2012
Efficient and accurate interactive image segmentation have significant importance in many multimedia applications. For mobile touchscreen-based applications, efficiency is more crucial. Moreover, due to small screens of the mobile devices, error tolerance is also a crucial factor. In this paper, a method for interactive image segmentation, tailored for mobile touch screen devices, is proposed. As an interaction methodology, coloring is presented. An automatic stroke-error correction methodology to correct the inaccurate user interaction is also proposed. For the efficient computation of the solution, a novel dynamic and iterative graph-cut solution is formulated. Efficiency and error tolerance of the proposed method are tested by using various sample images. Subjective evaluation of the interactive segmentation algorithms for mobile-touch screen is also performed. Indeed, for the challenging examples, the superior performance of the proposed method is obtained by the experiments.