Mean shift: An information theoretic perspective

Authors:
Sudhir Rao;Allan de Medeiros Martins;José C. Príncipe
Affiliations:
Computational Neuroengineering Laboratory (CNEL), Department of ECE, University of Florida, Gainesville, FL 32611, USA;Department of Automation and Computer Engineering Technology Center, Federal University of Rio Grande do Norte, Nadal, RN 59078-900, Brazil;Computational Neuroengineering Laboratory (CNEL), Department of ECE, University of Florida, Gainesville, FL 32611, USA
Venue:
Pattern Recognition Letters
Year:
2009

Citing 8
Cited 6

Normalized Cuts and Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Mode-Finding for Mixtures of Gaussian Distributions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Mean Shift: A Robust Approach Toward Feature Space Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
Mean Shift, Mode Seeking, and Clustering

IEEE Transactions on Pattern Analysis and Machine Intelligence
Mean Shift Is a Bound Optimization

IEEE Transactions on Pattern Analysis and Machine Intelligence
Fast nonparametric clustering with Gaussian blurring mean-shift

ICML '06 Proceedings of the 23rd international conference on Machine learning
Gaussian Mean-Shift Is an EM Algorithm

IEEE Transactions on Pattern Analysis and Machine Intelligence
The estimation of the gradient of a density function, with applications in pattern recognition

IEEE Transactions on Information Theory

Clustering using elements of information theory

ICANN'10 Proceedings of the 20th international conference on Artificial neural networks: Part III
An experimental study of color-based segmentation algorithms based on the mean-shift concept

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part II
Comparative study on information theoretic clustering and classical clustering algorithms

ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part II
Information-theoretic clustering: A representative and evolutionary approach

Expert Systems with Applications: An International Journal
Representative cross information potential clustering

Pattern Recognition Letters
Regularized discriminant entropy analysis

Pattern Recognition

Quantified Score

Hi-index	0.10

Visualization

Abstract

This paper develops a new understanding of mean shift algorithms from an information theoretic perspective. We show that the Gaussian blurring mean shift (GBMS) directly minimizes the Renyi's quadratic entropy of the dataset and hence is unstable by definition. Further, its stable counterpart, the Gaussian mean shift (GMS), minimizes the Renyi's ''cross'' entropy where the local stationary solutions are modes of the dataset. By doing so, we aptly answer the question ''What does mean shift algorithms optimize?'', thus highlighting naturally the properties of these algorithms. A consequence of this new understanding is the superior performance of GMS over GBMS which we show in a wide variety of applications ranging from mode finding to clustering and image segmentation.