A compressive sensing approach for Expression-invariant face recognition

Pradeep Nagesh, Baoxin Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

76 Citations (Scopus)

Abstract

We propose a novel technique based on compressive sensing for expression-invariant face recognition. We view the different images of the same subject as an ensemble of intercorrelated signals and assume that changes due to variation in expressions are sparse with respect to the whole image. We exploit this sparsity using distributed compressive sensing theory, which enables us to grossly represent the training images of a given subject by only two feature images: one that captures the holistic (common) features of the face, and the other that captures the different expressions in all training samples. We show that a new test image of a subject can be fairly well approximated using only the two feature images from the same subject. Hence we can drastically reduce the storage space and operational dimensionality by keeping only these two feature images or their random measurements. Based on this, we design an efficient expression-invariant classifier. Furthermore, we show that substantially low dimensional versions of the training features, such as (i) ones extracted from critically-downsampled training images, or (ii) low-dimensional random projection of original feature images, still have sufficient information for good classification. Extensive experiments with publically-available databases show that, on average, our approach performs better than the state-of-the-art despite using only such super-compact feature representation.

Original languageEnglish (US)
Title of host publication2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
Pages1518-1525
Number of pages8
DOIs
StatePublished - 2009
Event2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009 - Miami, FL, United States
Duration: Jun 20 2009Jun 25 2009

Other

Other2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
CountryUnited States
CityMiami, FL
Period6/20/096/25/09

Fingerprint

Face recognition
Classifiers
Experiments

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Biomedical Engineering

Cite this

Nagesh, P., & Li, B. (2009). A compressive sensing approach for Expression-invariant face recognition. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009 (pp. 1518-1525). [5206657] https://doi.org/10.1109/CVPRW.2009.5206657

A compressive sensing approach for Expression-invariant face recognition. / Nagesh, Pradeep; Li, Baoxin.

2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. 2009. p. 1518-1525 5206657.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nagesh, P & Li, B 2009, A compressive sensing approach for Expression-invariant face recognition. in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009., 5206657, pp. 1518-1525, 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009, Miami, FL, United States, 6/20/09. https://doi.org/10.1109/CVPRW.2009.5206657
Nagesh P, Li B. A compressive sensing approach for Expression-invariant face recognition. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. 2009. p. 1518-1525. 5206657 https://doi.org/10.1109/CVPRW.2009.5206657
Nagesh, Pradeep ; Li, Baoxin. / A compressive sensing approach for Expression-invariant face recognition. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. 2009. pp. 1518-1525
@inproceedings{e1155f34dd9e408693f31e7f27ea5e6c,
title = "A compressive sensing approach for Expression-invariant face recognition",
abstract = "We propose a novel technique based on compressive sensing for expression-invariant face recognition. We view the different images of the same subject as an ensemble of intercorrelated signals and assume that changes due to variation in expressions are sparse with respect to the whole image. We exploit this sparsity using distributed compressive sensing theory, which enables us to grossly represent the training images of a given subject by only two feature images: one that captures the holistic (common) features of the face, and the other that captures the different expressions in all training samples. We show that a new test image of a subject can be fairly well approximated using only the two feature images from the same subject. Hence we can drastically reduce the storage space and operational dimensionality by keeping only these two feature images or their random measurements. Based on this, we design an efficient expression-invariant classifier. Furthermore, we show that substantially low dimensional versions of the training features, such as (i) ones extracted from critically-downsampled training images, or (ii) low-dimensional random projection of original feature images, still have sufficient information for good classification. Extensive experiments with publically-available databases show that, on average, our approach performs better than the state-of-the-art despite using only such super-compact feature representation.",
author = "Pradeep Nagesh and Baoxin Li",
year = "2009",
doi = "10.1109/CVPRW.2009.5206657",
language = "English (US)",
isbn = "9781424439935",
pages = "1518--1525",
booktitle = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009",

}

TY - GEN

T1 - A compressive sensing approach for Expression-invariant face recognition

AU - Nagesh, Pradeep

AU - Li, Baoxin

PY - 2009

Y1 - 2009

N2 - We propose a novel technique based on compressive sensing for expression-invariant face recognition. We view the different images of the same subject as an ensemble of intercorrelated signals and assume that changes due to variation in expressions are sparse with respect to the whole image. We exploit this sparsity using distributed compressive sensing theory, which enables us to grossly represent the training images of a given subject by only two feature images: one that captures the holistic (common) features of the face, and the other that captures the different expressions in all training samples. We show that a new test image of a subject can be fairly well approximated using only the two feature images from the same subject. Hence we can drastically reduce the storage space and operational dimensionality by keeping only these two feature images or their random measurements. Based on this, we design an efficient expression-invariant classifier. Furthermore, we show that substantially low dimensional versions of the training features, such as (i) ones extracted from critically-downsampled training images, or (ii) low-dimensional random projection of original feature images, still have sufficient information for good classification. Extensive experiments with publically-available databases show that, on average, our approach performs better than the state-of-the-art despite using only such super-compact feature representation.

AB - We propose a novel technique based on compressive sensing for expression-invariant face recognition. We view the different images of the same subject as an ensemble of intercorrelated signals and assume that changes due to variation in expressions are sparse with respect to the whole image. We exploit this sparsity using distributed compressive sensing theory, which enables us to grossly represent the training images of a given subject by only two feature images: one that captures the holistic (common) features of the face, and the other that captures the different expressions in all training samples. We show that a new test image of a subject can be fairly well approximated using only the two feature images from the same subject. Hence we can drastically reduce the storage space and operational dimensionality by keeping only these two feature images or their random measurements. Based on this, we design an efficient expression-invariant classifier. Furthermore, we show that substantially low dimensional versions of the training features, such as (i) ones extracted from critically-downsampled training images, or (ii) low-dimensional random projection of original feature images, still have sufficient information for good classification. Extensive experiments with publically-available databases show that, on average, our approach performs better than the state-of-the-art despite using only such super-compact feature representation.

UR - http://www.scopus.com/inward/record.url?scp=70450186137&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450186137&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2009.5206657

DO - 10.1109/CVPRW.2009.5206657

M3 - Conference contribution

AN - SCOPUS:70450186137

SN - 9781424439935

SP - 1518

EP - 1525

BT - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009

ER -