A confidence-based late fusion framework for audio-visual biometric identification

Mohammad Alam, Mohammed Bennamoun, Roberto Togneri, Ferdous Sohel

    Research output: Contribution to journal › Article

    16 Citations (Scopus)
    296 Downloads (Pure)

    Abstract

    © 2014 Elsevier Inc. All rights reserved. This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio (C-ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches.
    Original languageEnglish
    Pages (from-to)65-71
    JournalPattern Recognition Letters
    Volume52
    DOIs
    Publication statusPublished - 15 Jan 2015

    Fingerprint

    Biometrics
    Fusion reactions
    Experiments

    Cite this

    Alam, Mohammad ; Bennamoun, Mohammed ; Togneri, Roberto ; Sohel, Ferdous. / A confidence-based late fusion framework for audio-visual biometric identification. In: Pattern Recognition Letters. 2015 ; Vol. 52. pp. 65-71.
    @article{a8e358dbf8db4d388cf81cc89a31aefc,
    title = "A confidence-based late fusion framework for audio-visual biometric identification",
    abstract = "{\circledC} 2014 Elsevier Inc. All rights reserved. This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio (C-ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches.",
    author = "Mohammad Alam and Mohammed Bennamoun and Roberto Togneri and Ferdous Sohel",
    year = "2015",
    month = "1",
    day = "15",
    doi = "10.1016/j.patrec.2014.10.006",
    language = "English",
    volume = "52",
    pages = "65--71",
    journal = "Pattern Recogniton Letters",
    issn = "0167-8655",
    publisher = "Elsevier",

    }

    Alam, M, Bennamoun, M, Togneri, R & Sohel, F 2015, 'A confidence-based late fusion framework for audio-visual biometric identification' Pattern Recognition Letters, vol. 52, pp. 65-71. https://doi.org/10.1016/j.patrec.2014.10.006

    A confidence-based late fusion framework for audio-visual biometric identification. / Alam, Mohammad; Bennamoun, Mohammed; Togneri, Roberto; Sohel, Ferdous.

    In: Pattern Recognition Letters, Vol. 52, 15.01.2015, p. 65-71.

    Research output: Contribution to journal › Article

    TY - JOUR

    T1 - A confidence-based late fusion framework for audio-visual biometric identification

    AU - Alam, Mohammad

    AU - Bennamoun, Mohammed

    AU - Togneri, Roberto

    AU - Sohel, Ferdous

    PY - 2015/1/15

    Y1 - 2015/1/15

    N2 - © 2014 Elsevier Inc. All rights reserved. This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio (C-ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches.

    AB - © 2014 Elsevier Inc. All rights reserved. This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio (C-ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches.

    U2 - 10.1016/j.patrec.2014.10.006

    DO - 10.1016/j.patrec.2014.10.006

    M3 - Article

    VL - 52

    SP - 65

    EP - 71

    JO - Pattern Recogniton Letters

    JF - Pattern Recogniton Letters

    SN - 0167-8655

    ER -

    Alam M, Bennamoun M, Togneri R, Sohel F. A confidence-based late fusion framework for audio-visual biometric identification. Pattern Recognition Letters. 2015 Jan 15;52:65-71. https://doi.org/10.1016/j.patrec.2014.10.006

    Access to Document

      Related content

      Grants

      Development of a 3D Audio Visual Next Generation Speech Recognition System

      Project: Research