Anisotropic Agglomerative Adaptive Mean-Shift

  title={Anisotropic Agglomerative Adaptive Mean-Shift},
  author={R. Sawhney and H. Christensen and G. Bradski},
  • R. Sawhney, H. Christensen, G. Bradski
  • Published 2014
  • Computer Science
  • ArXiv
  • Mean Shift today, is widely used for mode detection and clustering. The technique though, is challenged in practice due to assumptions of isotropicity and homoscedasticity. We present an adaptive Mean Shift methodology that allows for full anisotropic clustering, through unsupervised local bandwidth selection. The bandwidth matrices evolve naturally, adapting locally through agglomeration, and in turn guiding further agglomeration. The online methodology is practical and effecive for low… CONTINUE READING


    Publications referenced by this paper.
    Mean Shift, Mode Seeking, and Clustering
    • Y. Cheng
    • Mathematics, Computer Science
    • 1995
    • 3,326
    • PDF
    Mean Shift: A Robust Approach Toward Feature Space Analysis
    • 10,701
    • Highly Influential
    • PDF
    Quick Shift and Kernel Methods for Mode Seeking
    • 685
    • PDF
    Agglomerative Mean-Shift Clustering
    • 41
    • PDF
    Mean shift spectral clustering
    • 52
    • PDF
    Mean shift based clustering in high dimensions: a texture classification example
    • 482
    • Highly Influential
    • PDF
    An Algorithm for Data-Driven Bandwidth Selection
    • 384
    • Highly Influential
    • PDF
    A Topological Approach to Hierarchical Segmentation using Mean Shift
    • 192
    • PDF
    Fast nonparametric clustering with Gaussian blurring mean-shift
    • 104
    • PDF
    Gaussian Mean-Shift Is an EM Algorithm
    • 165
    • PDF