2D and 3D face recognition using convolutional neural network

Huiying Hu, Syed Afaq Ali Shah, Mohammed Bennamoun, Michael Molton

    Research output: Chapter in Book/Conference paperConference paper

    2 Citations (Scopus)

    Abstract

    Face recognition remains a challenge today as recognition performance is strongly affected by variability such as illumination, expressions and poses. In this work we apply Convolutional Neural Networks (CNNs) on the challenging task of both 2D and 3D face recognition. We constructed two CNN models, namely CNN-1 (two convolutional layers) and CNN-2 (one convolutional layer) for testing on 2D and 3D dataset. A comprehensive parametric study of two CNN models on face recognition is represented in which different combinations of activation function, learning rate and filter size are investigated. We find that CNN-2 has a better accuracy performance on both 2D and 3D face recognition. Our experimental results show that an accuracy of 85.15% was accomplished using CNN-2 on depth images with FRGCv2.0 dataset (4950 images with 557 objectives). An accuracy of 95% was achieved using CNN-2 on 2D raw image with the AT&T dataset (400 images with 40 objectives). The results indicate that the proposed CNN model is capable to handle complex information from facial images in different dimensions. These results provide valuable insights into further application of CNN on 3D face recognition.

    Original languageEnglish
    Title of host publicationTENCON 2017 - 2017 IEEE Region 10 Conference
    Place of PublicationUSA
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages133-138
    Number of pages6
    Volume2017-December
    ISBN (Electronic)9781509011339
    DOIs
    Publication statusPublished - 19 Dec 2017
    Event2017 IEEE Region 10 Conference, TENCON 2017 - Penang, Malaysia
    Duration: 5 Nov 20178 Nov 2017

    Conference

    Conference2017 IEEE Region 10 Conference, TENCON 2017
    CountryMalaysia
    CityPenang
    Period5/11/178/11/17

    Fingerprint

    Face recognition
    Neural networks
    Lighting
    Chemical activation
    Testing

    Cite this

    Hu, H., Shah, S. A. A., Bennamoun, M., & Molton, M. (2017). 2D and 3D face recognition using convolutional neural network. In TENCON 2017 - 2017 IEEE Region 10 Conference (Vol. 2017-December, pp. 133-138). USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/TENCON.2017.8227850
    Hu, Huiying ; Shah, Syed Afaq Ali ; Bennamoun, Mohammed ; Molton, Michael. / 2D and 3D face recognition using convolutional neural network. TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 133-138
    @inproceedings{4d8653bfb79f415caae432e0d75f188e,
    title = "2D and 3D face recognition using convolutional neural network",
    abstract = "Face recognition remains a challenge today as recognition performance is strongly affected by variability such as illumination, expressions and poses. In this work we apply Convolutional Neural Networks (CNNs) on the challenging task of both 2D and 3D face recognition. We constructed two CNN models, namely CNN-1 (two convolutional layers) and CNN-2 (one convolutional layer) for testing on 2D and 3D dataset. A comprehensive parametric study of two CNN models on face recognition is represented in which different combinations of activation function, learning rate and filter size are investigated. We find that CNN-2 has a better accuracy performance on both 2D and 3D face recognition. Our experimental results show that an accuracy of 85.15{\%} was accomplished using CNN-2 on depth images with FRGCv2.0 dataset (4950 images with 557 objectives). An accuracy of 95{\%} was achieved using CNN-2 on 2D raw image with the AT&T dataset (400 images with 40 objectives). The results indicate that the proposed CNN model is capable to handle complex information from facial images in different dimensions. These results provide valuable insights into further application of CNN on 3D face recognition.",
    keywords = "Convolutional Neural Networks, Depth Image, Face Recognition",
    author = "Huiying Hu and Shah, {Syed Afaq Ali} and Mohammed Bennamoun and Michael Molton",
    year = "2017",
    month = "12",
    day = "19",
    doi = "10.1109/TENCON.2017.8227850",
    language = "English",
    volume = "2017-December",
    pages = "133--138",
    booktitle = "TENCON 2017 - 2017 IEEE Region 10 Conference",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States",

    }

    Hu, H, Shah, SAA, Bennamoun, M & Molton, M 2017, 2D and 3D face recognition using convolutional neural network. in TENCON 2017 - 2017 IEEE Region 10 Conference. vol. 2017-December, IEEE, Institute of Electrical and Electronics Engineers, USA, pp. 133-138, 2017 IEEE Region 10 Conference, TENCON 2017, Penang, Malaysia, 5/11/17. https://doi.org/10.1109/TENCON.2017.8227850

    2D and 3D face recognition using convolutional neural network. / Hu, Huiying; Shah, Syed Afaq Ali; Bennamoun, Mohammed; Molton, Michael.

    TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 133-138.

    Research output: Chapter in Book/Conference paperConference paper

    TY - GEN

    T1 - 2D and 3D face recognition using convolutional neural network

    AU - Hu, Huiying

    AU - Shah, Syed Afaq Ali

    AU - Bennamoun, Mohammed

    AU - Molton, Michael

    PY - 2017/12/19

    Y1 - 2017/12/19

    N2 - Face recognition remains a challenge today as recognition performance is strongly affected by variability such as illumination, expressions and poses. In this work we apply Convolutional Neural Networks (CNNs) on the challenging task of both 2D and 3D face recognition. We constructed two CNN models, namely CNN-1 (two convolutional layers) and CNN-2 (one convolutional layer) for testing on 2D and 3D dataset. A comprehensive parametric study of two CNN models on face recognition is represented in which different combinations of activation function, learning rate and filter size are investigated. We find that CNN-2 has a better accuracy performance on both 2D and 3D face recognition. Our experimental results show that an accuracy of 85.15% was accomplished using CNN-2 on depth images with FRGCv2.0 dataset (4950 images with 557 objectives). An accuracy of 95% was achieved using CNN-2 on 2D raw image with the AT&T dataset (400 images with 40 objectives). The results indicate that the proposed CNN model is capable to handle complex information from facial images in different dimensions. These results provide valuable insights into further application of CNN on 3D face recognition.

    AB - Face recognition remains a challenge today as recognition performance is strongly affected by variability such as illumination, expressions and poses. In this work we apply Convolutional Neural Networks (CNNs) on the challenging task of both 2D and 3D face recognition. We constructed two CNN models, namely CNN-1 (two convolutional layers) and CNN-2 (one convolutional layer) for testing on 2D and 3D dataset. A comprehensive parametric study of two CNN models on face recognition is represented in which different combinations of activation function, learning rate and filter size are investigated. We find that CNN-2 has a better accuracy performance on both 2D and 3D face recognition. Our experimental results show that an accuracy of 85.15% was accomplished using CNN-2 on depth images with FRGCv2.0 dataset (4950 images with 557 objectives). An accuracy of 95% was achieved using CNN-2 on 2D raw image with the AT&T dataset (400 images with 40 objectives). The results indicate that the proposed CNN model is capable to handle complex information from facial images in different dimensions. These results provide valuable insights into further application of CNN on 3D face recognition.

    KW - Convolutional Neural Networks

    KW - Depth Image

    KW - Face Recognition

    UR - http://www.scopus.com/inward/record.url?scp=85044210743&partnerID=8YFLogxK

    U2 - 10.1109/TENCON.2017.8227850

    DO - 10.1109/TENCON.2017.8227850

    M3 - Conference paper

    VL - 2017-December

    SP - 133

    EP - 138

    BT - TENCON 2017 - 2017 IEEE Region 10 Conference

    PB - IEEE, Institute of Electrical and Electronics Engineers

    CY - USA

    ER -

    Hu H, Shah SAA, Bennamoun M, Molton M. 2D and 3D face recognition using convolutional neural network. In TENCON 2017 - 2017 IEEE Region 10 Conference. Vol. 2017-December. USA: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 133-138 https://doi.org/10.1109/TENCON.2017.8227850