Shape Description of 3D Objects by Curvature Spin Images Generated via Gaze Modeling

• Nakamae Takashi

  Kyushu Institute of Technology

• Maeda Makoto

  Kyushu Institute of Technology

• Inoue Katsuhiro

  Kyushu Institute of Technology

Abstract

To realize a model-based 3D object recognition, we propose a feature extraction method and a shape descriptor using the geometric features. First, the feature extraction method based on a novel gaze modeling is proposed. In the modeling process, the surface model is independently estimated for a part of range data restricted by several gaze domains. Hence, since the features are independently extracted from each gaze domain, inconsistent or incorrect features may be obtained. Therefore a stochastic method that enables us to integrate such features by evaluating the reliability of each gaze model is introduced. Next a shape descriptor, curvature spin image, is proposed. The CSI is created based on the ratio of surface curvatures. The main contribution of this paper is experimental analysis of the use of CSIs with various tuning parameters.

Journal

• Proceedings of the ISCIE International Symposium on Stochastic Systems Theory and its Applications

  Proceedings of the ISCIE International Symposium on Stochastic Systems Theory and its Applications 2011 (0), 196-201, 2011-05-05

  The ISCIE Symposium on Stochastic Systems Theory and Its Applications

Keywords

• Materials Science (miscellaneous)

Details

• CRID: 1390564237987199744
• NII Article ID: 130007377381
• DOI: 10.5687/sss.2011.196
• ISSN: 21884749
• ISSN: 21884730
• Web Site: https://www.jstage.jst.go.jp/article/sss/2011/0/2011_196/_pdf
• Text Lang: en
• Data Source:
  • JaLC
  • Crossref
  • CiNii Articles
• Abstract License Flag: Disallowed