Analysis, 3D reconstruction, & Animation of Faces

Abstract : Face analysis fields are widely spread out over a large quantity of domains : Human Computer Interaction, security, movies post-production, games... It includes detection, recognition, 3D reconstruction, animation, and emotion analysis. Face animation was the main motivation of this thesis, and we always kept it in mind at anytime. We discuss here, most of the fields. We first talk about face reconstruction and face modeling with a new model inspired by the Candide Model. Then, we address face detection and particularly features detection introducing anthropometric constraints in a global scheme through a Markov Random Field formulation. From prior constraints, we are able to estimate the 3d pose of a face from a single image and to extend it to motion tracking.We conlude our work with emotion analysis. We propose an expression modelling technique defined as time series, and we present our 3D database for face expression. We present the state of the art of emotion recognition. And we finally invoke the use of our expression modelling technique as a prediction to be compared with the data in time.
Document type :
Complete list of metadatas
Contributor : Vivien Girard <>
Submitted on : Wednesday, January 12, 2011 - 2:27:08 PM
Last modification on : Thursday, July 5, 2018 - 2:29:03 PM
Long-term archiving on : Wednesday, April 13, 2011 - 2:51:15 AM


  • HAL Id : pastel-00555140, version 1


Charlotte Ghys. Analysis, 3D reconstruction, & Animation of Faces. Human-Computer Interaction [cs.HC]. Ecole des Ponts ParisTech, 2010. English. ⟨NNT : 2010ENPC1005⟩. ⟨pastel-00555140⟩



Record views


Files downloads