MOMI-Cosegmentation: Simultaneous Segmentation of Multiple Objects among Multiple Images

Wen-Sheng Chu, Chia-Ping Chen, Chu-Song Chen

Abstract: In this study, we introduce a new cosegmentation approach, MOMI-cosegmentation, to segment multiple objects that repeatedly appear among multiple images. The proposed approach tackles a more general problem than conventional cosegmentation methods. Each of the shared objects may even appear more than one time in one image. The key idea of MOMI-cosegmentation is to incorporate a common pattern discovery algorithm with the proposed Gibbs energy model in a Markov random field framework. Our approach builds upon an observation that the detected common patterns provide useful information for estimating foreground statistics, while background statistics can be estimated from the remaining pixels. The initialization and segmentation processes of MOMI-cosegmentation are completely automatic, while the segmentation errors can be substantially reduced at the same time. Experimental results demonstrate the effectiveness of the proposed approach over state-of-the-art cosegmentation method.