Welcome to the Facial Image Processing and Analysis Group

The FIPA’s objective is to conduct research on automatic processing of facial images in order to build systems that can read the faces as humans do. The following research topics are covered by the group:

Face detection
Facial feature localization
Face modeling
Face recognition and verification
Face re-identification

Facial expression analysis
Emotion classification
Facial gesture recognition
Age estimation
Gender classification

The FIPA is directed by Dr.-Ing. Ekenel, who has several years of expertise on face analysis with the focus face recognition. He has received the EBF European Biometric Research Award in 2008 and, with his team, the Best Demo Award at the IEEE International Conference on Automatic Face and Gesture Recognition in 2008.

The FIPA is part of the Computer Vision for Human-Computer Interaction Lab in the Institute for Anthropomatics, at the Computer Science Department of the Karlsruhe Institute of Technology (KIT). The group is funded by the “Concept for the Future” of Karlsruhe Institute of Technology within the framework of the German Excellence Initiative.

NEWS

2nd BeFIT workshop at ECCV 2012

This year, we will organize the Second International Workshop on Benchmarking Facial Image Analysis Technologies (BeFIT 2012) in conjunction with ECCV 2012.

BeFIT Workshop 2012 website

FIPA has moved to its new office

Since April the 19th the FIPA group has moved to its new office in building 07.08, Vincenz-Priessnitz-Str. 3. We are located on the ground floor (EG) (have a look at OpenStreetMap or the interactive campus plan).

BeFIT workshop at ICCV 2011

Together with IBM Research and CAS , we organized the International Workshop on Benchmarking Facial Image Analysis Technologies (BeFIT 2011) in conjunction with ICCV 2011 .

BeFIT Workshop 2011 website

FIPA participated in the FERA2011 challenge

We participated in the emotion recognition and action unit detection tasks of the FG2011 Facial Expression Recognition and Analysis Challenge (FERA2011). We ranked 3rd in both tasks.

FERA2011 website