A Robust Spatio-temporal Face Modelling Approach Using 3D Multimodal Fusion for Biometric Security Applications

Girija Chetty, Michael Wagner

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

Abstract

In this paper, we propose a robust spatio-temporal face modelling approach based on multilevel fusion strategy involving cascaded fusion of hybrid multimodal fusion of audio-lip-face motion, correlation and depth features for biometric security application. The proposed approach combines the information from different audio-video based modules, namely: audio-lip motion module, audio-lip correlation module, 2D+3D motion-depth fusion module, and performs a hybrid cascaded fusion in an automatic, unsupervised and adaptive manner, by adapting to the local performance of each module. This is done by taking the output-score based reliability estimates (confidence measures) of each of the module into account. The module weightings are determined automatically such that the reliability measure of the combined scores is maximised. To test the robustness of the proposed approach, the audio and visual speech (mouth) modalities are degraded to emulate various levels of train/test mismatch; employing additive white Gaussian noise for the audio and JPEG compression for the video signals. The results show improved fusion performance for a range of tested levels of audio and video degradation, compared to the individual module performances. Experiments on a 3D stereovision database AVOZES show that, at severe levels of audio and video mismatch, the audio, mouth, 3D face, and tri-module (audio-lip motion, correlation and depth) fusion EERs were 42.9%, 32%, 15%, and 7.3% respectively for biometric identity verification scenario
Original languageEnglish
Title of host publicationProceedings of the SPIE, Biometric Technology for Human Identification V
EditorsB.V.K Vijaya Kumar, Salil Prabhakar, Arun Ross
Place of PublicationWashington
PublisherSPIE
Pages1-5
Number of pages5
Volume6944
DOIs
Publication statusPublished - 2008
EventSPIE Conference on Biometric Technology for Human Identification V - Orlando, United States
Duration: 18 Mar 200819 Mar 2008

Publication series

NameProceedings of SPIE
PublisherSPIE
Volume6944
ISSN (Print)0277-786X

Conference

ConferenceSPIE Conference on Biometric Technology for Human Identification V
CountryUnited States
CityOrlando
Period18/03/0819/03/08

Fingerprint Dive into the research topics of 'A Robust Spatio-temporal Face Modelling Approach Using 3D Multimodal Fusion for Biometric Security Applications'. Together they form a unique fingerprint.

  • Cite this

    Chetty, G., & Wagner, M. (2008). A Robust Spatio-temporal Face Modelling Approach Using 3D Multimodal Fusion for Biometric Security Applications. In B. V. K. V. Kumar, S. Prabhakar, & A. Ross (Eds.), Proceedings of the SPIE, Biometric Technology for Human Identification V (Vol. 6944, pp. 1-5). (Proceedings of SPIE; Vol. 6944). SPIE. https://doi.org/10.1117/12.778631