Semi-supervised kernel mean shift clustering

TY - JOUR

T1 - Semi-supervised kernel mean shift clustering

AU - Anand, Saket

AU - Mittal, Sushil

AU - Tuzel, Oncel

PY - 2014/6

Y1 - 2014/6

N2 - Mean shift clustering is a powerful nonparametric technique that does not require prior knowledge of the number of clusters and does not constrain the shape of the clusters. However, being completely unsupervised, its performance suffers when the original distance metric fails to capture the underlying cluster structure. Despite recent advances in semi-supervised clustering methods, there has been little effort towards incorporating supervision into mean shift. We propose a semi-supervised framework for kernel mean shift clustering (SKMS) that uses only pairwise constraints to guide the clustering procedure. The points are first mapped to a high-dimensional kernel space where the constraints are imposed by a linear transformation of the mapped points. This is achieved by modifying the initial kernel matrix by minimizing a log det divergence-based objective function. We show the advantages of SKMS by evaluating its performance on various synthetic and real datasets while comparing with state-of-the-art semi-supervised clustering algorithms.

AB - Mean shift clustering is a powerful nonparametric technique that does not require prior knowledge of the number of clusters and does not constrain the shape of the clusters. However, being completely unsupervised, its performance suffers when the original distance metric fails to capture the underlying cluster structure. Despite recent advances in semi-supervised clustering methods, there has been little effort towards incorporating supervision into mean shift. We propose a semi-supervised framework for kernel mean shift clustering (SKMS) that uses only pairwise constraints to guide the clustering procedure. The points are first mapped to a high-dimensional kernel space where the constraints are imposed by a linear transformation of the mapped points. This is achieved by modifying the initial kernel matrix by minimizing a log det divergence-based objective function. We show the advantages of SKMS by evaluating its performance on various synthetic and real datasets while comparing with state-of-the-art semi-supervised clustering algorithms.

KW - Semi-supervised kernel clustering

KW - log det Bregman divergence

KW - mean shift clustering

UR - http://www.scopus.com/inward/record.url?scp=84901821060&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901821060&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/TPAMI.2013.190

DO - https://doi.org/10.1109/TPAMI.2013.190

M3 - Article

SN - 0162-8828

VL - 36

SP - 1201

EP - 1215

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

IS - 6

M1 - 6616555

ER -