TY - GEN
T1 - Evolving mean shift with adaptive bandwidth
T2 - 9th Asian Conference on Computer Vision, ACCV 2009
AU - Zhao, Qi
AU - Yang, Zhi
AU - Tao, Hai
AU - Liu, Wentai
PY - 2010
Y1 - 2010
N2 - This paper presents a novel nonparametric clustering algorithm called evolving mean shift (EMS) algorithm. The algorithm iteratively shrinks a dataset and generates well formed clusters in just a couple of iterations. An energy function is defined to characterize the compactness of a dataset and we prove that the energy converges to zero at an exponential rate. The EMS is insensitive to noise as it automatically handles noisy data at an early stage. The single but critical user parameter, i.e., the kernel bandwidth, of the mean shift clustering family is adaptively updated to accommodate the evolving data density and alleviate the contradiction between global and local features. The algorithm has been applied and tested with image segmentation and neural spike sorting, where the improved accuracy can be obtained at a much faster performance, as demonstrated both qualitatively and quantitatively.
AB - This paper presents a novel nonparametric clustering algorithm called evolving mean shift (EMS) algorithm. The algorithm iteratively shrinks a dataset and generates well formed clusters in just a couple of iterations. An energy function is defined to characterize the compactness of a dataset and we prove that the energy converges to zero at an exponential rate. The EMS is insensitive to noise as it automatically handles noisy data at an early stage. The single but critical user parameter, i.e., the kernel bandwidth, of the mean shift clustering family is adaptively updated to accommodate the evolving data density and alleviate the contradiction between global and local features. The algorithm has been applied and tested with image segmentation and neural spike sorting, where the improved accuracy can be obtained at a much faster performance, as demonstrated both qualitatively and quantitatively.
UR - http://www.scopus.com/inward/record.url?scp=78650472868&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78650472868&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-12307-8_24
DO - 10.1007/978-3-642-12307-8_24
M3 - Conference contribution
AN - SCOPUS:78650472868
SN - 3642123066
SN - 9783642123061
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 258
EP - 268
BT - Computer Vision, ACCV 2009 - 9th Asian Conference on Computer Vision, Revised Selected Papers
Y2 - 23 September 2009 through 27 September 2009
ER -