A quantitative evaluation for 3D face reconstruction algorithms

In this work, we proposed to use quantitative method to evaluate the accuracy of 3D face reconstruction algorithms. The reconstructed 3D faces are first aligned to the ground truth by Iterative Closest Point (ICP) algorithm and then the shape difference between the two 3D faces is described by Signal to Noise Ratio (SNR). Finally, the error maps (EM) illustrated the reconstruction errors on corresponded vertices in different dimensions. Comparing with the subjective and indirect evaluation methods, the proposed method provides more precise and detailed evaluations for face shape reconstruction. Based on the SNR, different 3D face reconstruction algorithms can be compared directly and the EM also can suggest guidance for feature extraction.

[1]  Luc Van Gool,et al.  Modeling and synthesis of realistic visual speech in 3D , 2004 .

[2]  Christian Wallraven,et al.  Learning from humans: Computational modeling of face recognition , 2004, Network.

[3]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  P. Phillips,et al.  1 FACE RECOGNITION VENDOR TEST 2002 : EVALUATION REPORT , 2003 .

[5]  Yuxiao Hu,et al.  Real-time conversion from a single 2D face image to a 3D text-driven emotive audio-visual avatar , 2008, 2008 IEEE International Conference on Multimedia and Expo.

[6]  Xun Xu,et al.  Building Large Scale 3D Face Database for Face Analysis , 2007, MCAM.

[7]  Yuxiao Hu,et al.  Automatic 3D reconstruction for face recognition , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..