<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>6236</id>
    <completedYear/>
    <publishedYear>2016</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Detection and Characterization of Intrinsic Symmetry of 3D Shapes</title>
    <abstract language="eng">A comprehensive framework for detection and characterization&#13;
of partial intrinsic symmetry over 3D shapes is&#13;
proposed. To identify prominent symmetric regions which overlap&#13;
in space and vary in form, the proposed framework is decoupled&#13;
into a Correspondence Space Voting (CSV) procedure followed&#13;
by a Transformation Space Mapping (TSM) procedure. In the&#13;
CSV procedure, significant symmetries are first detected by&#13;
identifying surface point pairs on the input shape that exhibit&#13;
local similarity in terms of their intrinsic geometry while simultaneously&#13;
maintaining an intrinsic distance structure at a global&#13;
level. To allow detection of potentially overlapping symmetric&#13;
shape regions, a global intrinsic distance-based voting scheme is&#13;
employed to ensure the inclusion of only those point pairs that&#13;
exhibit significant intrinsic symmetry. In the TSM procedure,&#13;
the Functional Map framework is employed to generate the final&#13;
map of symmetries between point pairs. The TSM procedure&#13;
ensures the retrieval of the underlying dense correspondence map&#13;
throughout the 3D shape that follows a particular symmetry. The&#13;
TSM procedure is also shown to result in the formulation of a&#13;
metric symmetry space where each point in the space represents&#13;
a specific symmetry transformation and the distance between&#13;
points represents the complexity between the corresponding&#13;
transformations. Experimental results show that the proposed&#13;
framework can successfully analyze complex 3D shapes that&#13;
possess rich symmetries.</abstract>
    <parentTitle language="eng">Proceedings of IEEE International Conference on Pattern Recognition</parentTitle>
    <enrichment key="PeerReviewed">Yes</enrichment>
    <author>Anirban Mukhopadhyay</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Fatih Porikli</author>
    <author>Suchendra Bhandarkar</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
</export-example>
