Image Level Fusion Method for Multimodal 2D + 3D Face Recognition

Most of the existing multimodal 2D + 3D face recognition approaches do not account for the dependency between 2D and 3D representations of a face. This dependency reduces the benefit of fusion at the late-stage feature or metric level. On the other hand, it is advantageous to fuse at the early stage. We propose an image-level fusion method that explores the dependency between modalities for face recognition. Facial cues from 2D and 3D images are fused into more independent and discriminating data by finding fusion axes that pass through the most uncorrelated information in the images. Experimental results based on our face database of 1280 2D + 3D facial samples from 80 adults show that our image-level fusion approach outperforms the pixel- and metric-level fusion approaches.

[1]  Patrick J. Flynn,et al.  A survey of approaches and challenges in 3D and multi-modal 3D + 2D face recognition , 2006, Comput. Vis. Image Underst..

[2]  Chin-Seng Chua,et al.  Face recognition from 2D and 3D images using 3D Gabor filters , 2005, Image Vis. Comput..

[3]  Jacob Cohen.  Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[4]  Patrick J. Flynn,et al.  An evaluation of multimodal 2D+3D face biometrics , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Christoph von der Malsburg,et al.  Strategies and Benefits of Fusion of 2D and 3D Face Recognition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[6]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[7]  Alexander M. Bronstein,et al.  Expression-Invariant 3D Face Recognition , 2003, AVBPA.

[8]  Masato Nakajima,et al.  Face identification using a 3D gray-scale image-a method for lessening restrictions on facial directions , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.

[9]  Chiraz BenAbdelkader,et al.  Comparing and combining depth and texture cues for face recognition , 2005, Image Vis. Comput..

[10]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[11]  Michael G. Strintzis,et al.  Use of depth and colour eigenfaces for face recognition , 2003, Pattern Recognit. Lett..

[12]  Chin-Seng Chua,et al.  Facial feature detection and face recognition from 2D and 3D images , 2002, Pattern Recognit. Lett..

[13]  Marc Acheroy,et al.  Face verification from 3D and grey level clues , 2001, Pattern Recognit. Lett..

[14]  Andrea F. Abate,et al.  2D and 3D face recognition: A survey , 2007, Pattern Recognit. Lett..

[15]  Afzal Godil,et al.  Face recognition using 3D facial shape and color map information: comparison and combination , 2004, SPIE Defense + Commercial Sensing.

[16]  Anil K. Jain,et al.  Integrating Range and Texture Information for 3D Face Recognition , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[17]  Michael G. Strintzis,et al.  Face localization and authentication using color and depth images , 2005, IEEE Transactions on Image Processing.