Simplicity of linear representations (of images) makes them a popular tool in imaging analysis applications such as object recognition and image classification. Although several linear representations, namely PCA (principal component analysis), ICA, and FDA (Fisher discriminant analysis), have frequently been used, these representations are generally far from optimal in terms of actual application performance. We argue that representations should be chosen with respect to the application and the databases involved. Fixing an application, say object recognition, and assuming that recognition performance is computable for any linear basis (given a classifier and a database), we propose a Monte Carlo simulated annealing method that leads to optimal linear representations by maximizing the recognition performance over all fixed-rank subspaces. We illustrate this method on two popular databases.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Optimal linear representations of images for object recognition


    Contributors:
    Xiuwen Liu, (author) / Srivastava, A. (author) / Gallivan, K. (author)


    Publication date :

    2003-01-01


    Size :

    335104 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Optimal Linear Representations of Images for Object Recognition

    Liu, X. / Srivastava, A. / Gallivan, K. et al. | British Library Conference Proceedings | 2003


    Statistical Search for Hierarchical Linear Optimal Representations of Images

    Zhang, Qiang / Liu, Xiuwen / Srivastava, Anuj | IEEE | 2003


    Object-based representations of spatial images

    Newsam, S. / Bhagavathy, S. / Kenney, C. et al. | British Library Conference Proceedings | 2001


    Object-based representations of spatial images

    Newsam, Shawn | Online Contents | 2001


    Generic object recognition using multiple representations

    Das, S. / Bhanu, B. / Ho, C.-C. | British Library Online Contents | 1996