Unsupervised co-segmentation of a set of shapes via descriptor-space spectral clustering

Abstract

We introduce an algorithm for unsupervised co-segmentation of a set of shapes so as to reveal the semantic shape parts and establish their correspondence across the set. The input set may exhibit significant shape variability where the shapes do not admit proper spatial alignment and the corresponding parts in any pair of shapes may be geometrically dissimilar. Our algorithm can handle such challenging input sets since, first, we perform co-analysis in a descriptor space, where a combination of shape descriptors relates the parts independently of their pose, location, and cardinality. Secondly, we exploit a key enabling feature of the input set, namely, dissimilar parts may be "linked" through third-parties present in the set. The links are derived from the pairwise similarities between the parts' descriptors. To reveal such linkages, which may manifest themselves as anisotropic and non-linear structures in the descriptor space, we perform spectral clustering with the aid of diffusion maps. We show that with our approach, we are able to co-segment sets of shapes that possess significant variability, achieving results that are close to those of a supervised approach.

Active Contours with Group Similarity

“…Matrix factorization was used in these works to extract structure or motion information from detected feature points. Another topic closely related to our work is shape analysis such as shape clustering [26,25]. The difference is that sample shapes are provided as inputs in shape analysis while they are the outputs of segmentation.…”

Recovering Functional Mechanical Assemblies from Raw Scans

“…To recover the functional mechanical assembly from several raw scans, it is necessary to identify all the functional parts that are involved in the motion of the mechanical tool. As far as we know, purely geometric-based segmentation or co-segmentation methods [26], [27] are not sufficient for our purpose, which often involve training processes in order to account for semantics. On the other hand, as our input is raw scans under different motion configurations, which typically inhabit no correspondence information, thus deformation-based segmentation methods [24], [28] also do not work for our case.…”

Semantic Segmentation of Human Model Using Heat Kernel and Geodesic Distance

“…In recent years, the research on cosegmentation [24][25][26] has gradually become the hotspot of model segmentation. But in this paper, we do not relate to the content of this research.…”