Class-sensitive shape dissimilarity metric.

Shape dissimilarity is a fundamental problem with many applications such as shape exploration, retrieval, and classification. Given a collection of shapes, existing methods develop a consistent global metric to compare and organize shapes. The global nature of the involved shape descriptors implies that overall shape appearance is compared. These methods work well to distinguish shapes from different categories, but often fail for fine-grained classes within the same category. In this paper, we develop a dissimilarity metric for fine-grained classes by fusing together multiple distinctive metrics for different classes. The fused metric measures the dissimilarities among inter-class shapes by observing their unique traits. We demonstrate the advantage of using our approach in several applications.