Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Robust Topological Features for Deformation Invariant Image Matching

Edgar Lobaton, Ram Vasudevan, Ron Alterovitz and Ruzena Bajcsy

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2011-89
August 5, 2011

http://www.eecs.berkeley.edu/Pubs/TechRpts/2011/EECS-2011-89.pdf

Local photometric descriptors are a crucial low level component of numerous computer vision algorithms. In practice, these descriptors are constructed to be invariant to a class of transformations. However, the development of a descriptor that is simultaneously robust to noise and invariant under general deformation has proven difficult. In this paper, we introduce the Topological-Attributed Relational Graph (T-ARG), a new local photometric descriptor constructed from homology that is provably invariant to locally bounded deformation. This new robust topological descriptor is backed by a formal mathematical framework. We apply T-ARG to a set of benchmark images to evaluate its performance. Results indicate that T-ARG significantly outperforms traditional descriptors for noisy, deforming images.


BibTeX citation:

@techreport{Lobaton:EECS-2011-89,
    Author = {Lobaton, Edgar and Vasudevan, Ram and Alterovitz, Ron and Bajcsy, Ruzena},
    Title = {Robust Topological Features for Deformation Invariant Image Matching},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2011},
    Month = {Aug},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2011/EECS-2011-89.html},
    Number = {UCB/EECS-2011-89},
    Abstract = {Local photometric descriptors are a crucial low level component of numerous computer vision algorithms. In practice, these descriptors are constructed to be invariant to a class of transformations. However, the development of a descriptor that is simultaneously robust to noise and invariant under general deformation has proven difficult. In this paper, we introduce the Topological-Attributed Relational Graph (T-ARG), a new local photometric descriptor constructed from homology that is provably invariant to locally bounded deformation. This new robust topological descriptor is backed by a formal mathematical framework. We apply T-ARG to a set of benchmark images to evaluate its performance. Results indicate that T-ARG significantly outperforms traditional descriptors for noisy, deforming images.}
}

EndNote citation:

%0 Report
%A Lobaton, Edgar
%A Vasudevan, Ram
%A Alterovitz, Ron
%A Bajcsy, Ruzena
%T Robust Topological Features for Deformation Invariant Image Matching
%I EECS Department, University of California, Berkeley
%D 2011
%8 August 5
%@ UCB/EECS-2011-89
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2011/EECS-2011-89.html
%F Lobaton:EECS-2011-89