In this paper, we propose a novel method, called local non-negative matrix factorization (LNMF), for learning spatially localized, parts-based subspace representation of visual patterns. An objective function is defined to impose a localization constraint, in addition to the non-negativity constraint in the standard NMF. This gives a set of bases which not only allows a non-subtractive (part-based) representation of images but also manifests localized features. An algorithm is presented for the learning of such basic components. Experimental results are presented to compare LNMF with the NMF and PCA methods for face representation and recognition, which demonstrates advantages of LNMF.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Learning spatially localized, parts-based representation


    Beteiligte:
    Li, S.Z. (Autor:in) / Xin Wen Hou, (Autor:in) / Hong Jiang Zhang, (Autor:in) / Qian Sheng Cheng, (Autor:in)


    Erscheinungsdatum :

    2001-01-01


    Format / Umfang :

    728722 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Learning Spatially Localized, Parts-Based Representation

    Li, S. Z. / Hou, X. W. / Zhang, H. J. et al. | British Library Conference Proceedings | 2001


    Automatic View Recognition in Echocardiogram Videos Using Parts-Based Representation

    Ebadollahi, S. / Chang, S.-F. / Wu, H. et al. | British Library Conference Proceedings | 2004



    DTuD20 A localized representation of electromagnetic waves in photonic crystals

    Albert, J. P. / Jouanin, C. / Cassagne, D. | British Library Conference Proceedings | 2000


    Transient growth analysis of flow over an airfoil for identifying high-amplification, spatially-localized inputs

    Tamilselvam, Prashanth / Asztalos, Katherine J. / Dawson, Scott T. | AIAA | 2022