Mutual Learning Between Saliency and Similarity: Image Cosegmentation via Tree Structured Sparsity and Tree Graph Matching

Yan Ren, Licheng Jiao, Shuyuan Yang, Shuang Wang

Abstract: This paper proposes a unified mutual learning framework based on image hierarchies, which integrates structured sparsity with tree-graph matching to conquer the problem of weakly supervised image cosegmentation. We focus on the interaction between two common-object properties: saliency and similarity. Most existing cosegmentation methods only pay emphasis on either of them. The proposed method realizes the learning of the prior knowledge for structured sparsity with the help of tree-graph matching, which is capable of generating object-oriented salient regions. Meanwhile, it also reduces the searching space and computational complexity of tree-graph matching with the attendance of structured sparsity. We intend to thoughtfully exploit the hierarchically geometrical relationships of coherent objects. Experimental results compared with the state-of-the-arts on benchmark data sets confirm that the mutual learning framework is capable of effectively delineating co-existing object patterns in multiple images.