TY - GEN
T1 - Shrinkage expansion adaptive metric learning
AU - Wang, Qilong
AU - Zuo, Wangmeng
AU - Zhang, Lei
AU - Li, Peihua
PY - 2014/1/1
Y1 - 2014/1/1
N2 - Conventional pairwise constrained metric learning methods usually restrict the distance between samples of a similar pair to be lower than a fixed upper bound, and the distance between samples of a dissimilar pair higher than a fixed lower bound. Such fixed bound based constraints, however, may not work well when the intra- and inter-class variations are complex. In this paper, we propose a shrinkage expansion adaptive metric learning (SEAML) method by defining a novel shrinkage-expansion rule for adaptive pairwise constraints. SEAML is very effective in learning metrics from data with complex distributions. Meanwhile, it also suggests a new rule to assess the similarity between a pair of samples based on whether their distance is shrunk or expanded after metric learning. Our extensive experimental results demonstrated that SEAML achieves better performance than state-of-the-art metric learning methods. In addition, the proposed shrinkage-expansion adaptive pairwise constraints can be readily applied to many other pairwise constrained metric learning algorithms, and boost significantly their performance in applications such as face verification on LFW and PubFig databases.
AB - Conventional pairwise constrained metric learning methods usually restrict the distance between samples of a similar pair to be lower than a fixed upper bound, and the distance between samples of a dissimilar pair higher than a fixed lower bound. Such fixed bound based constraints, however, may not work well when the intra- and inter-class variations are complex. In this paper, we propose a shrinkage expansion adaptive metric learning (SEAML) method by defining a novel shrinkage-expansion rule for adaptive pairwise constraints. SEAML is very effective in learning metrics from data with complex distributions. Meanwhile, it also suggests a new rule to assess the similarity between a pair of samples based on whether their distance is shrunk or expanded after metric learning. Our extensive experimental results demonstrated that SEAML achieves better performance than state-of-the-art metric learning methods. In addition, the proposed shrinkage-expansion adaptive pairwise constraints can be readily applied to many other pairwise constrained metric learning algorithms, and boost significantly their performance in applications such as face verification on LFW and PubFig databases.
KW - adaptive bound constraints
KW - face verification
KW - pairwise constrained metric learning
KW - Shrinkage-expansion rule
UR - http://www.scopus.com/inward/record.url?scp=84906343437&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-10584-0_30
DO - 10.1007/978-3-319-10584-0_30
M3 - Conference article published in proceeding or book
SN - 9783319105833
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 456
EP - 471
BT - Computer Vision, ECCV 2014 - 13th European Conference, Proceedings
PB - Springer Verlag
T2 - 13th European Conference on Computer Vision, ECCV 2014
Y2 - 6 September 2014 through 12 September 2014
ER -