Virtual Mannequins Retrieval Based on Cluster-based Panoramic Views

Abstract

With the rapid development of 3D scanners, graphic accelerated hardware and modeling tools, the application of 3D model database is growing in both numbers and sizes, e.g., 3D head and body scan datasets. In this paper, a novel method by using cluster-based panoramic views is proposed for retrieving 3D virtual mannequins. First, a Young Chinese Female Virtual Mannequin (YCFVM) database is established by analyzing various 3D body scan data; second, panoramic views of virtual mannequins are obtained, and the Fourier and Wavelet features of the panoramic views are extracted; third, a multi-scale indexing of the YCFVM dataset is generated for fast matching by using fuzzy clustering for the Wavelet features; finally, the Fourier descriptor is embedded into the feature matching with the generated indexing for obtaining more accurate retrieval results. Experiment results show that the proposed method can achieve better retrieval performance than other descriptors used for virtual mannequin retrieval.