Anisotropic Agglomerative Adaptive Mean-Shift

Rahul Sawhney, Henrik Christensen and Gary Bradski

In Proceedings British Machine Vision Conference 2014


We present an adaptive Mean Shift methodology that allows full anisotropic clustering using 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 effective for low-dimensional feature spaces. Mean Shift today, is widely used for mode detection and clustering. In its popular fixed bandwidth isotropic form, it is known to be critically dependent and sensitive to bandwidth choice, on a per instance basis. The adaptive variants are isotropic and make use of inflexible heuristics; while offline selection methods ascertain a single global bandwidth and/or are data specific/non-automatic. The proposed method does away with the aforementioned limitations. A default form works well in general, preserving better detail and perceptual salience. The methodology though, allows for effective tuning of results.


Poster Session


Extended Abstract (PDF, 1 page, 2.8M)
Paper (PDF, 13 pages, 4.6M)
Supplemental Materials (ZIP, 196K)
Bibtex File


Rahul Sawhney, Henrik Christensen, and Gary Bradski. Anisotropic Agglomerative Adaptive Mean-Shift. Proceedings of the British Machine Vision Conference. BMVA Press, September 2014.


	title = {Anisotropic Agglomerative Adaptive Mean-Shift},
	author = {Sawhney, Rahul and Christensen, Henrik and Bradski, Gary},
	year = {2014},
	booktitle = {Proceedings of the British Machine Vision Conference},
	publisher = {BMVA Press},
	editors = {Valstar, Michel and French, Andrew and Pridmore, Tony}
	doi = { }