  Document Type  :  Latin Dissertation  Language of Document  :  English  Record Number  :  52574  Doc. No  :  TL22528  Call number  :  3295182  Main Entry  :  Mohammad Hossein Mahoor  Title & Author  :  A multimodal approach for face modeling and recognitionMohammad Hossein Mahoor  College  :  University of Miami  Date  :  2007  Degree  :  Ph.D.  student score  :  2007  Page No  :  170  Abstract  :  This dissertation describes a new methodology for multimodal (2D + 3D) face modeling and recognition. There are advantages in using each modality for face recognition. For example, the problems of pose variation and illumination condition, which cannot be resolved easily by using the 2D data, can be handled by using the 3D data. However, texture, which is provided by 2D data, is an important cue that cannot be ignored. Therefore, we use both the 2D and 3D modalities for face recognition and fuse the results of face recognition by each modality to boost the overall performance of the system. In this dissertation, we consider two different cases for multimodal face modeling and recognition. In the first case, the 2D and 3D data are registered. In this case we develop a unified graph model called Attributed Relational Graph (ARG) for face modeling and recognition. Based on the ARG model, the 2D and 3D data are included in a single model. The developed ARC model consists of nodes, edges, and mutual relations. The nodes of the graph correspond to the landmark points that are extracted by an improved Active Shape Model (ASM) technique. In order to extract the facial landmarks robustly, we improve the Active Shape Model technique by using the color information. Then, at each node of the graph, we calculate the response of a set of logGabor filters applied to the facial image texture and shape information (depth values); these features are used to model the local structure of the face at each node of the graph. The edges of the graph are defined based on Delaunay triangulation and a set of mutual relations between the sides of the triangles are defined. The mutual relations boost the final performance of the system. The results of face matching using the 2D and 3D attributes and the mutual relations are fused at the score level. In the second case, the 2D and 3D data are not registered. This lack of registration could be due to different reasons such as time lapse between the data acquisitions. Therefore, the 2D and 3D modalities are modeled independently. For the 3D modality, we developed a fully automated system for 3D face modeling and recognition based on ridge images. The problem with shape matching approaches such as Iterative Closest Points (ICP) or Hausdorff distance is the computational complexity. We model the face by 3D binary ridge images and use them for matching. In order to match the ridge points (either using the ICP or the Hausdorff distance), we extract three facial landmark points: namely, the two inner corners of the eyes and the tip of the nose, on the face surface using the Gaussian curvature. These three points are used for initial alignment of the constructed ridge images. As a result of using ridge points, which are just a fraction of the total points on the surface of the face, the computational complexity of the matching is reduced by two orders of magnitude. For the 2D modality, we model the face using an Attributed Relational Graph. The results of the 2D and 3D matching are fused at the score level. There are various techniques to fuse the 2D and 3D modalities. In this dissertation, we fuse the matching results at the score level to enhance the overall performance of our face recognition system. We compare the DempsterShafer theory of evidence and the weighted sum rule for fusion. We evaluate the performance of the above techniques for multimodal face recognition on various databases such as Gavab range database, FRGC (Face Recognition Grand Challenge) V2.0, and the University of Miami face database.  Subject  :  Applied sciences; Biometrics; Data fusion; Facial feature extraction; Multimodal face recognition; Range data; Threedimensional face recognition; Electrical engineering; Computer science; 0984:Computer science; 0544:Electrical engineering  Added Entry  :  M. AbdelMottaleb  Added Entry  :  University of Miami 
     
   
http://lib.clisel.com/site/catalogue/52574

 