Robust Topological Features For Deformation Invariant Image Matching

SpeakerDr. Edgar Lobaton
Organization NC State University
LocationEBII 1230
DateOctober 14, 2011 12:50 PM

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.

Dr. Edgar J. Lobaton received the Ph.D. in electrical engineering and computer sciences from the University of California, Berkeley in 2009. He completed his B.S. degrees in mathematics and electrical engineering from Seattle University in 2004. Dr. Lobaton is currently an Assistant Professor in the Department of Electrical and Computer Engineering at North Carolina State University. His areas of research include computer vision, sensor networks, robotics and control. He works on applications ranging from surveillance using smart camera systems to motion planning for medical robotic applications. Prior to joining NCSU, he was awarded the 2009 Computer Innovation Fellows post-doctoral fellowship award and conducted research in the Department of Computer Science at the University of North Carolina at Chapel Hill. He was also engaged in research at Alcatel-Lucent Bell Labs in 2005 and 2009.

  October 2011
Sun Mon Tues Wed Thu Fri Sat
      1
2345678
9101112131415
16171819202122
23242526272829
3031