Paper: Fast Graph Segmentation Based on Statistical Aggregation Phenomena

Fast Graph Segmentation Based on Statistical Aggregation Phenomena

Frank Nielsen and Richard Nock

Abstract:
In this paper, we first generalize a recent statistical color image segmentation algorithm [SRM'05] to arbitrary graphs, and report its performance for 2D images, 3D meshes and volume data. We then describe a fast pre-segmentation to the main graph procedure that allows us to further speed-up the segmentation by a factor of 2 to 4, without decreasing significantly the quality of segmentations. As an application, we built a real-time video segmentation that is robust enough to be used in camera-driven user interfaces and robotics applications.

Paper in PDF (4 pages, 1.5 MB)
Posters in PDF
Video of mesh segmentation
Video of real-time video segmentation

Java applet of statistical region merging
Java applet of statistical region merging with bias
Java applet of ClickRemoval, an application of the statistical region merging algorithm

BibTex entry:


@inproceedings{fsrm-2007,
	title = {Fast Graph Segmentation Based on Statistical Aggregation Phenomena},
	booktitle = {{ Proceedings of the 10th IAPR Conference on Machine Vision Applications }},
	author = { Frank Nielsen and Richard Nock },
	year = { 2007 },
	editor = { Katsushi Ikeuchi },
	location = { Institute of Industrial Science, The University of Tokyo, Institute of Industrial Science },
	publisher = { IAPR MVA Conference Committee },
	pages = {150--153},
	note = {ISBN 978-4-901122-07-8}
	address = { Tokyo, Japan }
}

Mesh segmentation:

Video segmentation:

3D medical image segmentation:

Last updated, 16th May 2007 5:00pm JT.