Efficient Fisher Discrimination Dictionary Learning
Jiang, Rui1; Qiao, Hong1,2; Zhang, Bo3,4; Qiao H(乔红)
AbstractFisher Determination Dictionary Learning (FDDL) has shown to be effective in image classification. However, the Original FDDL (O-FDDL) method is time-consuming. To address this issue, a fast Simplified FDDL (S-FDDL) method was proposed. But S-FDDL ignores the role of collaborative reconstruction, thus having an unstable performance in classification tasks with unbalanced changes in different classes. This paper focuses on developing an Efficient FDDL (E-FDDL) method, which is more suitable for such classification problems. Precisely, instead of solving the original Fisher Discrimination based Sparse Representation (FDSR) problem, we propose to solve an Approximate FDSR (A-FDSR) problem whose objective function is an upper bound of that of FDSR. A-FDSR considers the role of both the discriminative reconstruction and the collaborative reconstruction. This makes E-FDDL stable when dealing with classification tasks with unbalanced changes in different classes. Furthermore, fast optimization strategies are applicable to A-FDSR, thus leading to the high efficiency of E-FDDL which can be explained by analysis on convergence rate and computational complexity. We also use E-FDDL to accelerate the Shared Domain-adapted Dictionary Learning (SDDL) algorithm which is a FDDL based new method for domain adaptation. Experimental results on face and object recognition demonstrate the stable and fast performance of E-FDDL. (C) 2016 Elsevier B.V. All rights reserved.
Other Abstract
为了设计适用于不同类别变化不平衡的分类任务的快速Fisher判别性字典学习(Efficient Fisher Discrimination Dictionary Learning,E-FDDL)算法,我们提出解一个近似的Fisher判别性稀疏表示(Fisher Discrimination based Sparse Representation,FDSR)问题,它的目标函数是原始FDDL算法中FDSR问题目标函数的上界。该近似FDSR(Approximate FDSR,AFDSR)问题考虑了判别性重构和协同性重构两方面的作用,并且稍稍重视后者的作用,这使得E-FDDL在处理同类别变化不均匀的分类任务时更加鲁棒。进一步地,A-FDSR问题的结构使得快速的优化策略适用于该问题,这又带来了E-FDDL的快速性。我们在人脸识别实验的结果证实了E-FDDL快速稳定的表现。
KeywordFisher Discrimination Dictionary Learning Nesterov's Accelerated Gradient Method Face Recognition Domain Adaptation
WOS HeadingsScience & Technology ; Technology
Indexed BySCI
Funding OrganizationNational Natural Science Foundation of China(61210009 ; 61033011 ; 61379093 ; 11131006)
WOS Research AreaEngineering
WOS SubjectEngineering, Electrical & Electronic
WOS IDWOS:000379706500004
Citation statistics
Document Type期刊论文
Corresponding AuthorQiao H(乔红)
Affiliation1.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
2.Chinese Acad Sci, CAS Ctr Excellence Brain Sci & Intelligence Techn, Shanghai 200031, Peoples R China
3.Chinese Acad Sci, LSEC, Beijing 100190, Peoples R China
4.Chinese Acad Sci, AMSS, Inst Appl Math, Beijing 100190, Peoples R China
First Author AffilicationChinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Jiang, Rui,Qiao, Hong,Zhang, Bo,et al. Efficient Fisher Discrimination Dictionary Learning[J]. SIGNAL PROCESSING,2016,128(1):28-39.
APA Jiang, Rui,Qiao, Hong,Zhang, Bo,&乔红.(2016).Efficient Fisher Discrimination Dictionary Learning.SIGNAL PROCESSING,128(1),28-39.
MLA Jiang, Rui,et al."Efficient Fisher Discrimination Dictionary Learning".SIGNAL PROCESSING 128.1(2016):28-39.
Files in This Item: Download All
File Name/Size DocType Version Access License
SP-2016-Efficient Fi(2109KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Jiang, Rui]'s Articles
[Qiao, Hong]'s Articles
[Zhang, Bo]'s Articles
Baidu academic
Similar articles in Baidu academic
[Jiang, Rui]'s Articles
[Qiao, Hong]'s Articles
[Zhang, Bo]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Jiang, Rui]'s Articles
[Qiao, Hong]'s Articles
[Zhang, Bo]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: SP-2016-Efficient Fisher Discrimination Dictionary Learning-Final.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.