Perceptual Grouping via Untangling Gestalt Principles

Gestalt principles, a set of conjoining rules derived from hu- man visual studies, have been known to play an important role in computer vision. Many applications such as image segmentation, contour grouping and scene understanding of- ten rely on such rules to work. However, the problem of Gestalt confliction, i.e., the relative importance of each rule compared with another, remains unsolved. In this paper, we investigate the problem of perceptual grouping by quantifying the confliction among three commonly used rules: similarity, continuity and proximity. More specifically, we propose to quantify the importance of Gestalt rules by solving a learning to rank problem, and formulate a multi-label graph-cuts algo- rithm to group image primitives while taking into account the learned Gestalt confliction. Our experiment results confirm the existence of Gestalt confliction in perceptual grouping and demonstrate an improved performance when such a conflic- tion is accounted for via the proposed grouping algorithm. Finally, a novel cross domain image classification method is proposed by exploiting perceptual grouping as representation.