direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Face, Gesture and Deformations

Motivation & Project Description This project concerns mainly faces and deformable objects. Here is a brief list of topics related to the project: Face and gesture tracking and animation

  • Face modeling, detection and tracking
  • Face animation from video
  • Upper body tracking and gesture recognition
  • Facial model for speech and animation synthesis

Biometric identification

  • Iris authentication and identification
  • Face recognition

Physically based modeling

  • Physically valid models for tracking of highly flexible structures
  • Recovering forces and material properties from videos of deforming structures

Expected Outcome: The project is supposed to result in development of novel algorithms for face detection and pose estimation, face tracking and animation and face modelling. The face model for speech synthesis and animation should be developed. Equally project should provide advances in biometrics, especially developing new methods for iris recognition. Developing algorithms for upper body tracking and gesture recognition is yet another goal of the project. The project mainly should result in scientific publications and patents.

Time Frame: 

07/2007 - 07/2010

T-labs Team Members:
Slobodan Ilic, Nemanja Grujic, Rahul Swaminathan, Vladan Velisavljevic

CVLAB, EPFL, Switzerland / Technion, Israel / University of Valencia, Spain
Funding by:
Deutsche Telekom Laboratories
  • C. Strecha, H. Tyagi and P. Fua, Multi-class AdaBoost for face modelling, Technical Report, February 2010
  • V. Velisavljevic, Iris feature extraction, identification and verification system based on directionlets, EU Patent No. EP000002138950A1, filed June 2008, published December 2009
  • V. Velisavljevic, Low-complexity iris coding and recognition based on directionlets, IEEE Transactions on Information Forensics and Security, September 2009,
  • C. Strecha, A. Lindner, K. Ali and P. Fua, Training for task specific keypoint detection, Annual Symposium of the German Association for Pattern Recognition (DAGM), 2009
  • S. Ilic, Object labeling for recognition using vocabulary trees, IEEE International Conference in Pattern Recognition (ICPR), December 2008
  • N. Grujic, S. Ilic, V. Lepetit and P. Fua, 3D facial pose estimation by image retrieval, IEEE International Conference on Face and Gesture Recognition, September 2008
  • S.Ilic, P.Fua, Non-Linear Beam Model for Tracking Large Deformations.http://www.deutsche-telekom-laboratories.de/%7Eslobodan/publications/07/IlicFuaICCV07.pdf IEEE International Conference on Computer Vision, Rio de Janeiro, Brazil, October 2007
  • S.Ilic, Comparing Timoshenko Beam to Energy Beam for Fitting Noisy Data. Asian Converence in Computer Vision, Tokyo, Japan, November, 2007

Zusatzinformationen / Extras


Schnellnavigation zur Seite über Nummerneingabe