Contact: Dr.-Ing. Hazim K. Ekenel
Our aim is to build face recognition systems for real-life scenarios that:
- do not require elaborate face alignment,
- are less sensitive to illumination and background variations,
- can operate on both still images and video
- can detect & recognize unknown people
- are robust against face detection failures - partially detected faces, false detections.
Our ultimate goal is to build a face recognition system that can learn faces without any supervision and recognize people unawarely. Therefore, we are conducting research both at the representation and classification level to improve the performance of the face recognition systems:
- Representation Level: Local Appearance Based Face Recognition
In this study we are developing a generic local appearance based face representation using Discrete Cosine Transform (DCT).
- Classification Level: Two-Class Linear Discriminant Analysis for Face Recognition
In this approach we divide a single M-class linear discriminant classifier into M 2-class linear classifiers. This formulation of linear discriminant classifier has many advantages like simpler calculation of projection vectors, more discrimination between classes, easier update of the database with new individuals and inherent detection of unknown people.