• DocumentCode
    2959607
  • Title

    Integrating local classifiers through nonlinear dynamics on label graphs with an application to image segmentation

  • Author

    Chen, Yutian ; Gelfand, Andrew ; Fowlkes, Charless C. ; Welling, Max

  • Author_Institution
    Bren Sch. of Inf. & Comput. Sci., Univ. of California, Irvine, CA, USA
  • fYear
    2011
  • fDate
    6-13 Nov. 2011
  • Firstpage
    2635
  • Lastpage
    2642
  • Abstract
    We present a new method to combine possibly inconsistent locally (piecewise) trained conditional models p(yα|xα) into pseudo-samples from a global model. Our method does not require training of a CRF, but instead generates samples by iterating forward a weakly chaotic dynamical system. The new method is illustrated on image segmentation tasks where classifiers based on local appearance cues are combined with pairwise boundary cues.
  • Keywords
    graph theory; image classification; image segmentation; time-varying systems; conditional model; image segmentation; label graph; local classifier integration; nonlinear dynamics; pairwise boundary cues; weakly chaotic dynamical system; Accuracy; Computational modeling; Data models; Image segmentation; Joints; Mathematical model; Training;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Vision (ICCV), 2011 IEEE International Conference on
  • Conference_Location
    Barcelona
  • ISSN
    1550-5499
  • Print_ISBN
    978-1-4577-1101-5
  • Type

    conf

  • DOI
    10.1109/ICCV.2011.6126553
  • Filename
    6126553
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2959607