Scale-space clustering with recursive validation

Authors:
Tomoya Sakai;Takuto Komazaki;Atsushi Imiya
Affiliations:
Institute of Media and Information Technology, Chiba University, Japan;Graduate School of Science and Technology, Chiba University, Japan;Institute of Media and Information Technology, Chiba University, Japan
Venue:
SSVM'07 Proceedings of the 1st international conference on Scale space and variational methods in computer vision
Year:
2007

Citing 10
Cited 0

Determining number of clusters and prototype locations via multi-scale clustering

Pattern Recognition Letters
The Topological Structure of Scale-Space Images

Journal of Mathematical Imaging and Vision
Clustering by Scale-Space Filtering

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
Scale-Space Theory in Computer Vision

Scale-Space Theory in Computer Vision
Linear Scale-Space has First been Proposed in Japan

Journal of Mathematical Imaging and Vision
Scale Space Hierarchy

Journal of Mathematical Imaging and Vision
Scale-space filtering

IJCAI'83 Proceedings of the Eighth international joint conference on Artificial intelligence - Volume 2
Mode estimation using pessimistic scale space tracking

Scale Space'03 Proceedings of the 4th international conference on Scale space methods in computer vision
Scale-Space hierarchy of singularities

DSSCV'05 Proceedings of the First international conference on Deep Structure, Singularities, and Computer Vision

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a hierarchical clustering method for a dataset based on the deep structure of the probability density function (PDF) of the data in the scale space. The data clusters correspond to the modes of the PDF, and their hierarchy is determined by regarding the nonparametric estimation of the PDF with the Gaussian kernel as a scale-space representation. It is shown that the number of clusters is statistically deterministic above a certain critical scale, even though the positions of the data points are stochastic. Such a critical scale is estimated by analysing the distribution of cluster lifetime in the scale space, and statistically valid clusters are detected above the critical scale. This cluster validation using the critical scale can be recursively employed according to the hierarchy of the clusters.