Reconstruction of Human Face for Different Age Based on the Log-gabor Wavelet

A new method is proposed for reconstruction of different age human face based on the face shape and age texture. First, average face shapes are obtained by punctuating on different age faces, and the faces are warped by radial basis function to the average shapes. Second, by decomposing and selecting the elderly face images using the log-Gabor wavelet, we can get age texture, then add it to the warped face. Finally, we can reconstruct the target age of human face by judging the age of human face through the age estimation function, adjusting age texture parameter that can control wrinkle´s changes. Experimental results show that the method can reconstruct different age for human faces which are more sincerity.