Skip to Main content Skip to Navigation
Theses

Identity authentication through 3D face analysis

Abstract : This thesis presents a complete approach for face recognition that exploits the volume information and its robustness in order to reduce the influences of viewpoint and illumination on performances. The main contributions of this work concern facial surface acquisition and recognition. Facial surfaces are conveniently captured with the so-called structured light technique. A precision in depth of about 1 mm satisfies facial comparison needs. The acquisition speed and the natural cooperation of the user fulfil elementary conditions of a practical implementation. The recognition of facial surfaces is carried out by geometrical comparison of profiles extracted by parallel planes. The intrinsic symmetry of the face allowed for the optimisation of profile matching. The geometrical information was combined with the comparison of grey levels measured in the images acquired for 3D capture.
Document type :
Theses
Domain :
Complete list of metadatas

https://pastel.archives-ouvertes.fr/pastel-00000620
Contributor : Ecole Télécom Paristech <>
Submitted on : Friday, April 16, 2004 - 8:00:00 AM
Last modification on : Friday, July 31, 2020 - 10:44:07 AM
Long-term archiving on: : Thursday, September 30, 2010 - 6:28:03 PM

Identifiers

  • HAL Id : pastel-00000620, version 1

Collections

Citation

Charles Beumier. Identity authentication through 3D face analysis. domain_other. Télécom ParisTech, 2003. English. ⟨pastel-00000620⟩

Share

Metrics

Record views

385

Files downloads

315