The recently introduced random walker segmentation algorithm by Grady and Funka-Lea (2004) has been shown to have desirable theoretical properties and to perform well on a wide variety of images in practice. However, this algorithm requires user-specified labels and produces a segmentation where each segment is connected to a labeled pixel. We show that incorporation of a nonparametric probability density model allows for an extended random walkers algorithm that can locate disconnected objects and does not require user-specified labels. Finally, we show that this formulation leads to a deep connection with the popular graph cuts method by Boykov et al. (2001) and Wu and Leahy (1993).


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Multilabel random walker image segmentation using prior models


    Contributors:
    Grady, L. (author)


    Publication date :

    2005-01-01


    Size :

    360852 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Multilabel Spatial Image Recognition using Deep Convolutional Neural Network

    Bhat, Nagaraj / Archana Hebbar, K V / Bhat, Sachin et al. | IEEE | 2020


    An Image Segmentation Algorithm of Adaptive Prior MRF Models

    Liu, G. / Ren, Q. / Meng, Y. et al. | British Library Online Contents | 2013



    Multilabel partition moves for MRF optimization

    Shabou, A. / Darbon, J. / Tupin, F. | British Library Online Contents | 2013


    Joint Prior Models of Neighboring Objects for 3D Image Segmentation

    Yang, J. / Duncan, J. / IEEE Computer Society | British Library Conference Proceedings | 2004