Authors:
Farshad Nourbakhsh
and
Eric Granger
Affiliation:
Université du Québec, Canada
Keyword(s):
Matrix Factorization, Graph Compression, Dictionary Learning, Sparse Representation Classification, Face Recognition, Video Surveillance.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Biometrics
;
Biometrics and Pattern Recognition
;
Classification
;
Clustering
;
Computer Vision, Visualization and Computer Graphics
;
Image and Video Analysis
;
Image Understanding
;
Matrix Factorization
;
Multimedia
;
Multimedia Signal Processing
;
Object Recognition
;
Pattern Recognition
;
Software Engineering
;
Sparsity
;
Telecommunications
;
Theory and Methods
;
Video Analysis
Abstract:
Despite the limited target data available to design face models in video surveillance applications, many faces
of non-target individuals may be captured over multiple cameras in operational environments to improve robustness
to variations. This paper focuses on Sparse Representation Classification (SRC) techniques that are
suitable for the design of still-to-video FR systems based on under-sampled dictionaries. The limited reference
data available during enrolment is complemented by an over-complete external dictionary that is formed
with an abundance of faces from non-target individuals. In this paper, the Graph-Compressed Dictionary
Learning (GCDL) technique is proposed to learn compact auxiliary dictionaries for SRC. GCDL is based on
matrix factorization, and allows to maintain a high level of SRC accuracy with compressed dictionaries because
it exploits structural information to represent intra-class variations. Graph compression based on matrix
factorization shown to efficiently
compress data, and can therefore rapidly construct compact dictionaries. Accuracy
and efficiency of the proposed GCDL technique is assessed and compared to reference sparse coding
and dictionary learning techniques using images from the CAS-PEAL database. GCDL is shown to provide
fast matching and adaptation of compressed dictionaries to new reference faces from the video surveillance
environments.
(More)