Improving efficiency of density-based shape descriptors for 3D object retrieval. Akgul, C. B., Sankur, B., Yemez, Y., & Schmitt, F. In Gagalowicz, A & Philips, W, editors, Computer Vision/Computer Graphics Collaboration Techniques, volume 4418, of LECTURE NOTES IN COMPUTER SCIENCE, pages 330-340, 2007. INRA; Ghent Univ. 3rd International Conference on Computer Vision/Computer Graphics (MIRAGE 2007), Rocquencourt, FRANCE, MAR 28-30, 2007
abstract   bibtex   
We consider 3D shape description as a probability modeling problem. The local surface properties are first measured via various features, and then the probability density function (pdf) of the multidimensional feature vector becomes the shape descriptor. Our prior work has shown that, for 3D object retrieval, pdf-based schemes can provide descriptors that are computationally efficient and performance-wise on a par with or better than the state-of-the-art methods. In this paper, we specifically focus on discretization problems in the multidimensional feature space, selection of density evaluation points and dimensionality reduction techniques to further improve the performance of our density-based descriptors.
@inproceedings{ ISI:000246010500030,
Author = {Akgul, Ceyhun Burak and Sankur, Bulent and Yemez, Yiicel and Schmitt,
   Francis},
Editor = {{Gagalowicz, A and Philips, W}},
Title = {{Improving efficiency of density-based shape descriptors for 3D object
   retrieval}},
Booktitle = {{Computer Vision/Computer Graphics Collaboration Techniques}},
Series = {{LECTURE NOTES IN COMPUTER SCIENCE}},
Year = {{2007}},
Volume = {{4418}},
Pages = {{330-340}},
Note = {{3rd International Conference on Computer Vision/Computer Graphics
   (MIRAGE 2007), Rocquencourt, FRANCE, MAR 28-30, 2007}},
Organization = {{INRA; Ghent Univ}},
Abstract = {{We consider 3D shape description as a probability modeling problem. The
   local surface properties are first measured via various features, and
   then the probability density function (pdf) of the multidimensional
   feature vector becomes the shape descriptor. Our prior work has shown
   that, for 3D object retrieval, pdf-based schemes can provide descriptors
   that are computationally efficient and performance-wise on a par with or
   better than the state-of-the-art methods. In this paper, we specifically
   focus on discretization problems in the multidimensional feature space,
   selection of density evaluation points and dimensionality reduction
   techniques to further improve the performance of our density-based
   descriptors.}},
ISSN = {{0302-9743}},
ISBN = {{978-3-540-71456-9}},
Unique-ID = {{ISI:000246010500030}},
}

Downloads: 0