An Improved Accuracy Rate for Face Authentication with Pose Adjustment based-on 2D-3D Transformation

In face authentication and other face biometric methods, an image of a person can be misclassified if the pose of their face is different than that of the training data unless there are steps taken to eliminate these inaccuracies. The methods in this paper are designed to improve the accuracy of a face authentication system when the pose between the input image and training images are different. First, an efficient 2D-to-3D integrated face reconstruction approach is introduced to reconstruct a personalized 3D face model from a single frontal face image with neutral expression and normal illumination; Then, realistic virtual faces with different of pose are synthesized based on the personalized 3D face to characterize the face subspace; Finally, face recognition is conducted based on these representative virtual faces. Compared with other related works, this framework has the following advantages: 1) only one single frontal face is required for face recognition, which avoids the burdensome enrollment work; and, 2) the synthesized face samples provide the capability to conduct recognition under difficult conditions like complex pose, illumination and expression. From the experimental results show that the synthesized virtual faces significantly improve the accuracy of face recognition with variant pose, illumination and expression.