Shape-based retrieval and analysis of 3D models using fuzzy weighted symmetrical depth images

  • Authors:
  • Kuan-Sheng Zou;Chee-Kooi Chan;Si-Xiang Peng;Ameersing Luximon;Zeng-Qiang Chen;Wai-Hung Ip

  • Affiliations:
  • Department of Automation, Nankai University, Tianjin 300071, China and Institute of Textiles & Clothing, Hong Kong Polytechnic University, Hong Kong;Institute of Textiles & Clothing, Hong Kong Polytechnic University, Hong Kong;Institute of Textiles & Clothing, Hong Kong Polytechnic University, Hong Kong;Institute of Textiles & Clothing, Hong Kong Polytechnic University, Hong Kong;Department of Automation, Nankai University, Tianjin 300071, China;Department of Industrial and Systems Engineering, Hong Kong Polytechnic University, Hong Kong

  • Venue:
  • Neurocomputing
  • Year:
  • 2012

Quantified Score

Hi-index 0.01

Visualization

Abstract

With the rapid development of 3D scanners, graphic accelerated hardware and 3D modeling tools, the application of 3D model databases is growing in both numbers and sizes, e.g. 3D body scans, head model and virtual mannequins. There is a pressing need for effective content-based 3D model retrieval methods. In this paper, a novel 3D model retrieval approach is proposed by using the Fuzzy Weighted Symmetrical Depth Images (FW-SDI). Firstly, three symmetrical planes of the 3D model are obtained based upon principal plane analysis and sequential quadratic programming, which are the distinctive characteristics of 3D model and perpendicular to each other. Secondly, three novel depth Images (NDI) and three depth differential images (DDI) are extracted by projecting the 3D surface to the proposed symmetrical planes. The Fourier descriptors of the novel depth Images and depth differential images are calculated. Finally, a fuzzy weighted procedure is conducted for combining the Fourier descriptors of NDI and DDI. Experiment results show that the proposed method can achieve better retrieval performance than others.