ICIP 2006, Atlanta, GA
 

Slide Show

Atlanta Conv. & Vis. Bureau

 

Technical Program

Paper Detail

Paper:MP-P5.3
Session:Machine Learning for Image and Video Classification
Time:Monday, October 9, 14:20 - 17:00
Presentation: Poster
Title: LOCAL DISCRIMINANT EMBEDDING WITH TENSOR REPRESENTATION
Authors: Jian Xia; Hong Kong University of Science and Technology 
 Dit-Yan Yeung; Hong Kong University of Science and Technology 
 Guang Dai; Hong Kong University of Science and Technology 
Abstract: We present a subspace learning method, called Local Discriminant Embedding with Tensor representation (LDET), that addresses simultaneously the generalization and data representation problems in subspace learning. LDET learns multiple interrelated subspaces for obtaining a lower-dimensional embedding by incorporating both class label information and neighborhood information. By encoding each object as a second- or higher-order tensor, LDET can capture higher-order structures in the data without requiring a large sample size. Extensive empirical studies have been performed to compare LDET with a second- or third-order tensor representation and the original LDE on their face recognition performance. Not only does LDET have a lower computational complexity than LDE, but LDET is also superior to LDE in terms of its recognition accuracy.