DocumentCode :
3518185
Title :
Online codebook reweighting using pairwise constraints for image classification
Author :
Zhao, Xin ; Ren, Weiqiang ; Huang, Kaiqi ; Tan, Tieniu
Author_Institution :
Dept. of Autom., Univ. of Sci. & Technol. of China, Hefei, China
fYear :
2011
fDate :
28-28 Nov. 2011
Firstpage :
662
Lastpage :
666
Abstract :
Bag-of-words (BoW) model is widely used for image classification. Recently, the framework of sparse coding and max pooling proved an effective approach for image classification. Max pooling adopts a winner-take-all strategy. Thus, it can be regarded as a codebook weighting process. The results of this process are the weights of the associated codebook. However, there are high intra-class variations and strong background clutters in many image classification tasks. The weights obtained by max pooling only have limited information. This paper presents a codebook reweighting algorithm using pairwise constraints to improve the performance of sparse coding and max pooling framework. Pairwise constraints are the natural way of encoding the relationships between pairs of images. Therefore, the reweighted codebook is more effective to describe the relevance between pairs of images. An efficient online learning algorithm is presented based on passive-aggressive training strategy. We compare our method with other state-of-the-art methods on Graz-01 & 02 datasets. Experimental results illustrate the effectiveness and efficiency of our method for image classification.
Keywords :
clutter; image classification; image coding; learning (artificial intelligence); BoW model; background clutter; bag-of-words model; image classification; intraclass variations; max pooling framework; online codebook reweighting algorithm; online learning algorithm; pairwise constraints; passive-aggressive training strategy; sparse coding; winner-take-all strategy; Clutter; Encoding; Image coding; Image representation; Optimization; Training; Vectors; Codebook reweighting; Image classification; Online learning; Pairwise constraints;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (ACPR), 2011 First Asian Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4577-0122-1
Type :
conf
DOI :
10.1109/ACPR.2011.6166560
Filename :
6166560
Link To Document :
بازگشت