Dynamic biometrics fusion at feature level for video-based human recognition
Contribuinte(s) |
Cree, Michael J. |
---|---|
Data(s) |
01/01/2007
|
Resumo |
This paper proposes a novel human recognition method in video, which combines human face and gait traits<br />using a dynamic multi-modal biometrics fusion scheme. The Fisherface approach is adopted to extract face<br />features, while for gait features, Locality Preserving Projection (LPP) is used to achieve low-dimensional<br />manifold embedding of the temporal silhouette data derived from image sequences. Face and gait features are<br />fused dynamically at feature level based on a distance-driven fusion method. Encouraging experimental results<br />are achieved on the video sequences containing 20 people, which show that dynamically fused features produce<br />a more discriminating power than any individual biometric as well as integrated features built on common static<br />fusion schemes.<br /> |
Identificador | |
Idioma(s) |
eng |
Publicador |
Image and Vision Computing NZ |
Relação |
http://dro.deakin.edu.au/eserv/DU:30008097/geng-dynamicbiometricsfusion-2007.pdf http://digital.liby.waikato.ac.nz/conferences/ivcnz07/papers/ivcnz07-paper29.pdf |
Direitos |
2007, Image and Vision Computing NZ |
Palavras-Chave | #human recognition #multimodal biometrics #dynamic fusion |
Tipo |
Conference Paper |