Robust Face Recognition via Multi-Scale Patch-Based Matrix Regression

Autor: Jian Yang, Juliang Hua, Dong Yue, Xiao-Yuan Jing, Pu Huang, Guangwei Gao
Jazyk: angličtina
Rok vydání: 2016
Předmět:
Time Factors
Databases
Factual
Computer science
Social Sciences
lcsh:Medicine
02 engineering and technology
Facial recognition system
Pattern Recognition
Automated
Database and Informatics Methods
Matrix (mathematics)
Cognition
Learning and Memory
Law Enforcement
Medicine and Health Sciences
0202 electrical engineering
electronic engineering
information engineering
Psychology
lcsh:Science
Multidisciplinary
Applied Mathematics
Simulation and Modeling
Regression
Physical Sciences
020201 artificial intelligence & image processing
Anatomy
Algorithms
Research Article
Optimization
Imaging Techniques
Matrix norm
Research and Analysis Methods
Face Recognition
Human Learning
Memory
Robustness (computer science)
Learning
Humans
Computer Simulation
business.industry
lcsh:R
Cognitive Psychology
Biology and Life Sciences
020206 networking & telecommunications
Pattern recognition
Sample size determination
Face
Cognitive Science
Perception
Law and Legal Sciences
lcsh:Q
Artificial intelligence
business
Head
Mathematics
Criminal Justice System
Neuroscience
Zdroj: PLoS ONE, Vol 11, Iss 8, p e0159945 (2016)
PLoS ONE
ISSN: 1932-6203
Popis: In many real-world applications such as smart card solutions, law enforcement, surveillance and access control, the limited training sample size is the most fundamental problem. By making use of the low-rank structural information of the reconstructed error image, the so-called nuclear norm-based matrix regression has been demonstrated to be effective for robust face recognition with continuous occlusions. However, the recognition performance of nuclear norm-based matrix regression degrades greatly in the face of the small sample size problem. An alternative solution to tackle this problem is performing matrix regression on each patch and then integrating the outputs from all patches. However, it is difficult to set an optimal patch size across different databases. To fully utilize the complementary information from different patch scales for the final decision, we propose a multi-scale patch-based matrix regression scheme based on which the ensemble of multi-scale outputs can be achieved optimally. Extensive experiments on benchmark face databases validate the effectiveness and robustness of our method, which outperforms several state-of-the-art patch-based face recognition algorithms.
Databáze: OpenAIRE
Externí odkaz: