Research on skeleton matching classification

Abstract

Skeleton matching plays a very important role in image retrieval, which can retrieve the graphics corresponding to the graphics library by the matching degree of object skeleton. Therefore, the key step of shape retrieval is to perform effective bone matching. The skeleton can also be used for surface segmentation, especially when the two deformation models are divided into the same number of surface slices, the difference between the surface slices of the two models needs to be used to build a continuous map, which is the case. In this paper, the skeleton matching methods are summarized, and the advantages and disadvantages of each algorithm are described, which provides theoretical basis for target recognition, shape retrieval, medical image analysis and processing, augmented reality and other fields.