|
Ph.D Defense Thursday, September 20, 2007 ENG II – Room 216 10:00AM Title: Super-resolution and Facial Expressions for Face Recognition in Video Abstract: Face recognition based on video has received significant attention in the past few years. However, the facial images in video that is acquired from a distance are usually small and their quality is low. Enhancing low-resolution (LR) facial images from a video sequence is of importance for performing face recognition. Super-resolution (SR) reconstruction is one of the most difficult and ill-posed problems due to the demand of accurate alignments between multiple images and multiple solutions for a given set of images. In particular, human face is much more complex and elastic compared to other objects which are addressed by the majority of the super-resolution literature. Super-resolution from facial images may suffer from subtle facial expression variation, non-rigid complex motion, visibility and occlusion, illumination and reflectance variations. In this dissertation, the objective is to design algorithms to tackle the problems brought by the special characteristics of human face. The techniques presented in this dissertation address facial expression variations, non-rigidity of face and illumination changes. The key contributions are: 1) A closed-loop system for incremental super-resolution of video and its use for face recognition. The system uses a generic 3D model of the face and compensates for changing illumination and 3D pose in video. 2) A super-resolution system that explicitly accounts for facial expressions by treating the face as the composition of local face entities (eyes, nose, mouth, eyebrows and rest of the face) and performing appropriate distortions to the face. 3) A genetically-inspired learning method for facial expression recognition. Unlike most of the current research on facial expression recognition that generally selects visually meaningful feature by hand, the proposed learning method can discover the features automatically in a genetic programming-based approach. The results will be presented with a wide variety of video and images. |
| < Prev | Next > |
|---|