TY - GEN
T1 - Metric learning for semi-supervised clustering of region covariance descriptors
AU - Sivalingam, Ravishankar
AU - Morellas, Vassilios
AU - Boley, Daniel
AU - Papanikolopoulos, Nikolaos
PY - 2009
Y1 - 2009
N2 - in this paper we extend distance metric learning to a new class of descriptors known as Region Covariance Descriptors. Region covariances are becoming increasingly popular as features for object detection and classification over the past few years. Given a set of pairwise constraints by the user, we want to perform semi-supervised clustering of these descriptors aided by metric learning approaches. The covariance descriptors belong to the special class of symmetric positive definite (SPD) tensors, and current algorithms cannot deal with them directly without violating their positive definiteness. In our framework, the distance metric on the manifold of SPD matrices is represented as an L2 distance in a vector space, and a Mahalanobis-type distance metric is learnt in the new space, in order to improve the performance of semi-supervised clustering of region covariances. We present results from clustering of covariance descriptors representing different human images, from single and multiple camera views. This transformation from a set of positive definite tensors to a Euclidean space paves the way for the application of many other vector-space methods to this class of descriptors.
AB - in this paper we extend distance metric learning to a new class of descriptors known as Region Covariance Descriptors. Region covariances are becoming increasingly popular as features for object detection and classification over the past few years. Given a set of pairwise constraints by the user, we want to perform semi-supervised clustering of these descriptors aided by metric learning approaches. The covariance descriptors belong to the special class of symmetric positive definite (SPD) tensors, and current algorithms cannot deal with them directly without violating their positive definiteness. In our framework, the distance metric on the manifold of SPD matrices is represented as an L2 distance in a vector space, and a Mahalanobis-type distance metric is learnt in the new space, in order to improve the performance of semi-supervised clustering of region covariances. We present results from clustering of covariance descriptors representing different human images, from single and multiple camera views. This transformation from a set of positive definite tensors to a Euclidean space paves the way for the application of many other vector-space methods to this class of descriptors.
KW - Appearance clustering
KW - Distance metric learning
KW - Region covariance descriptors
KW - Semi-supervised clustering
UR - http://www.scopus.com/inward/record.url?scp=72149110053&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=72149110053&partnerID=8YFLogxK
U2 - 10.1109/ICDSC.2009.52893415
DO - 10.1109/ICDSC.2009.52893415
M3 - Conference contribution
AN - SCOPUS:72149110053
SN - 9781424446209
T3 - 2009 3rd ACM/IEEE International Conference on Distributed Smart Cameras, ICDSC 2009
BT - 2009 3rd ACM/IEEE International Conference on Distributed Smart Cameras, ICDSC 2009
T2 - 2009 3rd ACM/IEEE International Conference on Distributed Smart Cameras, ICDSC 2009
Y2 - 30 August 2009 through 2 September 2009
ER -