Illumination and expression invariant face recognition using SSIM based sparse representation

Asim Khwaja, Akshay Asthana, Roland Goecke

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    4 Citations (Scopus)

    Abstract

    The sparse representation technique has provided a new way of looking at object recognition. As we demonstrate in this paper, however, the mean-squared error (MSE) measure, which is at the heart of this technique, is not a very robust measure when it comes to comparing facial images, which differ significantly in luminance values, as it only performs pixel-by-pixel comparisons. This requires a significantly large training set with enough variations in it to offset the drawback of the MSE measure. A large training set, however, is often not available. We propose the replacement of the MSE measure by the structural similarity (SSIM) measure in the sparse representation algorithm, which performs a more robust comparison using only one training sample per subject. In addition, since the off-the-shelf sparsifiers are also written using the MSE measure, we developed our own sparsifier using genetic algorithms that use the SSIM measure. We applied the modified algorithm to the Extended Yale Face B database as well as to the Multi-PIE database with expression and illumination variations. The improved performance demonstrates the effectiveness of the proposed modifications.
    Original languageEnglish
    Title of host publication2010 20th International Conference on Pattern Recognition (ICPR)
    Place of PublicationPiscataway, NJ, U.S.A.
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages4028-4031
    Number of pages4
    ISBN (Print)9780769541099
    DOIs
    Publication statusPublished - 2010
    EventICPR 2010: 20th International Conference on Pattern Recognition - Istanbul, Turkey
    Duration: 23 Aug 201026 Aug 2010

    Conference

    ConferenceICPR 2010: 20th International Conference on Pattern Recognition
    CountryTurkey
    CityIstanbul
    Period23/08/1026/08/10

    Fingerprint

    Face recognition
    Lighting
    Pixels
    Object recognition
    Luminance
    Genetic algorithms

    Cite this

    Khwaja, A., Asthana, A., & Goecke, R. (2010). Illumination and expression invariant face recognition using SSIM based sparse representation. In 2010 20th International Conference on Pattern Recognition (ICPR) (pp. 4028-4031). Piscataway, NJ, U.S.A.: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICPR.2010.979
    Khwaja, Asim ; Asthana, Akshay ; Goecke, Roland. / Illumination and expression invariant face recognition using SSIM based sparse representation. 2010 20th International Conference on Pattern Recognition (ICPR). Piscataway, NJ, U.S.A. : IEEE, Institute of Electrical and Electronics Engineers, 2010. pp. 4028-4031
    @inproceedings{6b65df334a354de3bbf29b02a3176ebf,
    title = "Illumination and expression invariant face recognition using SSIM based sparse representation",
    abstract = "The sparse representation technique has provided a new way of looking at object recognition. As we demonstrate in this paper, however, the mean-squared error (MSE) measure, which is at the heart of this technique, is not a very robust measure when it comes to comparing facial images, which differ significantly in luminance values, as it only performs pixel-by-pixel comparisons. This requires a significantly large training set with enough variations in it to offset the drawback of the MSE measure. A large training set, however, is often not available. We propose the replacement of the MSE measure by the structural similarity (SSIM) measure in the sparse representation algorithm, which performs a more robust comparison using only one training sample per subject. In addition, since the off-the-shelf sparsifiers are also written using the MSE measure, we developed our own sparsifier using genetic algorithms that use the SSIM measure. We applied the modified algorithm to the Extended Yale Face B database as well as to the Multi-PIE database with expression and illumination variations. The improved performance demonstrates the effectiveness of the proposed modifications.",
    author = "Asim Khwaja and Akshay Asthana and Roland Goecke",
    year = "2010",
    doi = "10.1109/ICPR.2010.979",
    language = "English",
    isbn = "9780769541099",
    pages = "4028--4031",
    booktitle = "2010 20th International Conference on Pattern Recognition (ICPR)",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States",

    }

    Khwaja, A, Asthana, A & Goecke, R 2010, Illumination and expression invariant face recognition using SSIM based sparse representation. in 2010 20th International Conference on Pattern Recognition (ICPR). IEEE, Institute of Electrical and Electronics Engineers, Piscataway, NJ, U.S.A., pp. 4028-4031, ICPR 2010: 20th International Conference on Pattern Recognition, Istanbul, Turkey, 23/08/10. https://doi.org/10.1109/ICPR.2010.979

    Illumination and expression invariant face recognition using SSIM based sparse representation. / Khwaja, Asim; Asthana, Akshay; Goecke, Roland.

    2010 20th International Conference on Pattern Recognition (ICPR). Piscataway, NJ, U.S.A. : IEEE, Institute of Electrical and Electronics Engineers, 2010. p. 4028-4031.

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    TY - GEN

    T1 - Illumination and expression invariant face recognition using SSIM based sparse representation

    AU - Khwaja, Asim

    AU - Asthana, Akshay

    AU - Goecke, Roland

    PY - 2010

    Y1 - 2010

    N2 - The sparse representation technique has provided a new way of looking at object recognition. As we demonstrate in this paper, however, the mean-squared error (MSE) measure, which is at the heart of this technique, is not a very robust measure when it comes to comparing facial images, which differ significantly in luminance values, as it only performs pixel-by-pixel comparisons. This requires a significantly large training set with enough variations in it to offset the drawback of the MSE measure. A large training set, however, is often not available. We propose the replacement of the MSE measure by the structural similarity (SSIM) measure in the sparse representation algorithm, which performs a more robust comparison using only one training sample per subject. In addition, since the off-the-shelf sparsifiers are also written using the MSE measure, we developed our own sparsifier using genetic algorithms that use the SSIM measure. We applied the modified algorithm to the Extended Yale Face B database as well as to the Multi-PIE database with expression and illumination variations. The improved performance demonstrates the effectiveness of the proposed modifications.

    AB - The sparse representation technique has provided a new way of looking at object recognition. As we demonstrate in this paper, however, the mean-squared error (MSE) measure, which is at the heart of this technique, is not a very robust measure when it comes to comparing facial images, which differ significantly in luminance values, as it only performs pixel-by-pixel comparisons. This requires a significantly large training set with enough variations in it to offset the drawback of the MSE measure. A large training set, however, is often not available. We propose the replacement of the MSE measure by the structural similarity (SSIM) measure in the sparse representation algorithm, which performs a more robust comparison using only one training sample per subject. In addition, since the off-the-shelf sparsifiers are also written using the MSE measure, we developed our own sparsifier using genetic algorithms that use the SSIM measure. We applied the modified algorithm to the Extended Yale Face B database as well as to the Multi-PIE database with expression and illumination variations. The improved performance demonstrates the effectiveness of the proposed modifications.

    U2 - 10.1109/ICPR.2010.979

    DO - 10.1109/ICPR.2010.979

    M3 - Conference contribution

    SN - 9780769541099

    SP - 4028

    EP - 4031

    BT - 2010 20th International Conference on Pattern Recognition (ICPR)

    PB - IEEE, Institute of Electrical and Electronics Engineers

    CY - Piscataway, NJ, U.S.A.

    ER -

    Khwaja A, Asthana A, Goecke R. Illumination and expression invariant face recognition using SSIM based sparse representation. In 2010 20th International Conference on Pattern Recognition (ICPR). Piscataway, NJ, U.S.A.: IEEE, Institute of Electrical and Electronics Engineers. 2010. p. 4028-4031 https://doi.org/10.1109/ICPR.2010.979