2017
DOI: 10.1109/tip.2017.2675341
|View full text |Cite
|
Sign up to set email alerts
|

Semi-Supervised Sparse Representation Based Classification for Face Recognition With Insufficient Labeled Samples

Abstract: This paper addresses the problem of face recognition when there is only few, or even only a single, labeled examples of the face that we wish to recognize. Moreover, these examples are typically corrupted by nuisance variables, both linear (i.e., additive nuisance variables, such as bad lighting and wearing of glasses) and non-linear (i.e., non-additive pixel-wise nuisance variables, such as expression changes). The small number of labeled examples means that it is hard to remove these nuisance variables betwe… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
88
0

Year Published

2017
2017
2020
2020

Publication Types

Select...
5
4
1

Relationship

2
8

Authors

Journals

citations
Cited by 230 publications
(88 citation statements)
references
References 54 publications
0
88
0
Order By: Relevance
“…An early finding to compensate visual domain shift in FR systems is to employ a generic set to enrich the diversity of the reference gallery set that is the so-called generic learning concept [14]. Generic learning has been widely discussed by many researchers [15], [16]. Su et al proposed an adaptive generic learning method for FR which utilized external data to estimate the within-class scatter matrix for each individual and applies this information to the reference set [14].…”
Section: B Generic Learningmentioning
confidence: 99%
“…An early finding to compensate visual domain shift in FR systems is to employ a generic set to enrich the diversity of the reference gallery set that is the so-called generic learning concept [14]. Generic learning has been widely discussed by many researchers [15], [16]. Su et al proposed an adaptive generic learning method for FR which utilized external data to estimate the within-class scatter matrix for each individual and applies this information to the reference set [14].…”
Section: B Generic Learningmentioning
confidence: 99%
“…However, face recognition in low lighting condition is still a very challenging problem, e.g., the near infrared (NIR) image based face recognition [3]. In recent years, VIS-NIR heterogeneous face recognition has attracted more and more attention from researchers and enterprises due to its practical value in many real-world applications, e.g., security surveillance and E-passport [4], [5].…”
Section: Introductionmentioning
confidence: 99%
“…Sparse representation has recently been studied in a variety of problems [28][29][30]. Sparse unmixing is a semi-supervised unmixing method, which assumes that the observed HSI can be formulated to find the optimal subset of pure spectral signatures from a prior large spectral library.…”
Section: Introductionmentioning
confidence: 99%