*banner
 

Unsupervised Segmentation of Natural Images via Lossy Data Compression
Allen Yang, John Wright, Shankar Sastry, Yi Ma

Citation
Allen Yang, John Wright, Shankar Sastry, Yi Ma. "Unsupervised Segmentation of Natural Images via Lossy Data Compression". Technical report, University of California, Berkeley, UCB/EECS-2006-195, December, 2006.

Abstract
In this paper, we cast natural-image segmentation as a problem of clustering texture features as multivariate mixed data. We model the distribution of the texture features using a mixture of Gaussian distributions. However, unlike most existing clustering methods, we allow the mixture components to be degenerate or nearly-degenerate. We contend that this assumption is particularly important for mid-level image segmentation, where degeneracy is typically introduced by using a common feature representation for different textures. We show that such a mixture distribution can be effectively segmented by a simple agglomerative clustering algorithm derived from a lossy data compression approach. Using simple fixed-size Gaussian windows as texture features, the algorithm segments an image by minimizing the overall coding length of all the feature vectors. In terms of a variety of performance indices, our algorithm compares favorably against other well-known image segmentation methods on the Berkeley image database.

Electronic downloads

Citation formats  
  • HTML
    Allen Yang, John Wright, Shankar Sastry, Yi Ma. <a
    href="http://chess.eecs.berkeley.edu/pubs/220.html"
    ><i>Unsupervised Segmentation of Natural Images via
    Lossy Data Compression</i></a>, Technical
    report,  University of California, Berkeley,
    UCB/EECS-2006-195, December, 2006.
  • Plain text
    Allen Yang, John Wright, Shankar Sastry, Yi Ma.
    "Unsupervised Segmentation of Natural Images via Lossy
    Data Compression". Technical report,  University of
    California, Berkeley, UCB/EECS-2006-195, December, 2006.
  • BibTeX
    @techreport{YangWrightSastryMa06_UnsupervisedSegmentationOfNaturalImagesViaLossyDataCompression,
        author = {Allen Yang and John Wright and Shankar Sastry and
                  Yi Ma},
        title = {Unsupervised Segmentation of Natural Images via
                  Lossy Data Compression},
        institution = {University of California, Berkeley},
        number = {UCB/EECS-2006-195},
        month = {December},
        year = {2006},
        abstract = {In this paper, we cast natural-image segmentation
                  as a problem of clustering texture features as
                  multivariate mixed data. We model the distribution
                  of the texture features using a mixture of
                  Gaussian distributions. However, unlike most
                  existing clustering methods, we allow the mixture
                  components to be degenerate or nearly-degenerate.
                  We contend that this assumption is particularly
                  important for mid-level image segmentation, where
                  degeneracy is typically introduced by using a
                  common feature representation for different
                  textures. We show that such a mixture distribution
                  can be effectively segmented by a simple
                  agglomerative clustering algorithm derived from a
                  lossy data compression approach. Using simple
                  fixed-size Gaussian windows as texture features,
                  the algorithm segments an image by minimizing the
                  overall coding length of all the feature vectors.
                  In terms of a variety of performance indices, our
                  algorithm compares favorably against other
                  well-known image segmentation methods on the
                  Berkeley image database.},
        URL = {http://chess.eecs.berkeley.edu/pubs/220.html}
    }
    

Posted by Allen Yang on 24 Apr 2007.
Groups: chesslocal
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.

Notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright.

©2002-2018 Chess