Learning Distinctive Local Object Characteristics for 3D Shape Retrieval

O. Deussen, D. Keim und D. Saupe (Editoren)
In proceedings of Vision, Modeling, and Visualization 2008 (VMV 2008), pages 167-178, Akademische Verlagsgesellschaft Aka GmbH, Heidelberg, Okt. 2008
Präsentiert: The 8th International Fall Workshop Vision, Modeling and Visualisation 2008
 

Abstract

While supervised learning approaches for 3D shape retrieval have been successfully used to incorporate human knowledge about object classes based on global shape features, the incorporation of local features still remains a difficult task. First, it is not obvious how to measure the similarity between two objects each represented by a set of local features, and second, it is not clear how to choose local feature scales such that they are most distinctive. In this paper, we tackle both of these problems and present a supervised learning approach that uses arbitrary local features for 3D shape retrieval. It avoids the problem of establishing feature correspondences and automatically detects discriminating feature scales. Our experiments on the Princeton Shape Benchmark show that our method is superior to state-of-the-art shape retrieval techniques.

Bilder

Paper herunterladen

Paper herunterladen

Bibtex

@INPROCEEDINGS{wessel-2008-learning,
      author = {Wessel, Raoul and Baranowski, Rafael and Klein, Reinhard},
      editor = {Deussen, O. and Keim, D. and Saupe, D.},
       pages = {167--178},
       title = {Learning Distinctive Local Object Characteristics for 3D Shape Retrieval},
   booktitle = {Vision, Modeling, and Visualization 2008 (VMV 2008)},
        year = {2008},
       month = oct,
   publisher = {Akademische Verlagsgesellschaft Aka GmbH, Heidelberg},
    abstract = {While supervised learning approaches for 3D shape
                retrieval have been successfully used to incorporate
                human knowledge about object classes based
                on global shape features, the incorporation of local
                features still remains a difficult task. First, it is
                not obvious how to measure the similarity between
                two objects each represented by a set of local features,
                and second, it is not clear how to choose local
                feature scales such that they are most distinctive.
                In this paper, we tackle both of these problems
                and present a supervised learning approach
                that uses arbitrary local features for 3D shape retrieval.
                It avoids the problem of establishing feature
                correspondences and automatically detects discriminating
                feature scales. Our experiments on the
                Princeton Shape Benchmark show that our method
                is superior to state-of-the-art shape retrieval techniques.},
        isbn = {978-3-89838-609-8},
  conference = {The 8th International Fall Workshop Vision, Modeling and Visualisation 2008}
}