Detail publikace

3D Face Recognition

Originální název

3D Face Recognition

Anglický název

3D Face Recognition

Jazyk

en

Originální abstrakt

This work is about biometric 3D face recognition. General biometric systems as well as specific techniques used in 2D and 3D face recognition are described. An automatic modular 3D face recognition method is proposed. First, facial landmarks are located on the face surface and the three dimensional model is aligned to a predefined position. After that, the input probe scan is compared to the gallery template. There are three fundamental face recognition algorithms employed during this process - the eigenface method (PCA), the recognition using histogram-based features, and the recognition based on the anatomical-Bertillon face features. The decision module fuses the scores provided by the utilized recognition techniques finally. The resulting performance is better than any of utilized recognition algorithms.

Anglický abstrakt

This work is about biometric 3D face recognition. General biometric systems as well as specific techniques used in 2D and 3D face recognition are described. An automatic modular 3D face recognition method is proposed. First, facial landmarks are located on the face surface and the three dimensional model is aligned to a predefined position. After that, the input probe scan is compared to the gallery template. There are three fundamental face recognition algorithms employed during this process - the eigenface method (PCA), the recognition using histogram-based features, and the recognition based on the anatomical-Bertillon face features. The decision module fuses the scores provided by the utilized recognition techniques finally. The resulting performance is better than any of utilized recognition algorithms.

BibTex


@book{BUT91123,
  author="Štěpán {Mráček}",
  title="3D Face Recognition",
  annote="This work is about biometric 3D face recognition. General biometric systems as
well as specific techniques used in 2D and 3D face recognition are described. An
automatic modular 3D face recognition method is proposed. First, facial landmarks
are located on the face surface and the three dimensional model is aligned to
a predefined position. After that, the input probe scan is compared to the
gallery template. There are three fundamental face recognition algorithms
employed during this process - the eigenface method (PCA), the recognition using
histogram-based features, and the recognition based on the anatomical-Bertillon
face features. The decision module fuses the scores provided by the utilized
recognition techniques finally. The resulting performance is better than any of
utilized recognition algorithms.",
  address="Lambert Academic Publishing",
  chapter="91123",
  edition="NEUVEDEN",
  howpublished="print",
  institution="Lambert Academic Publishing",
  year="2011",
  month="november",
  pages="0--0",
  publisher="Lambert Academic Publishing",
  type="book"
}