基于Gabor小波的2D人脸表情的合成

人脸图像合成是新一代人机交互中的重要技术。传统的三维模型加生理模型的方法可以生成真实的人脸表情图像，但计算复杂度很高。文章提出一种基于Gabor小波的用于人脸逼真表情图像合成的新方法，利用Gabor小波变换滤波器提取人脸表情图像特征，生成逼真表情图像。从实验结果来看该方法可以仅利用1张训练集内、外的人脸图像合成出该人在不同表情下逼真的脸部表情图像，同时可以合成库内的人在新表情下的表情图像。

2-D Subtle Facial Expression Synthesis Based on Gabor Wavelet

Facial expression synthesis is an important technique inhuman computer interactions. The traditional 3-D reconstruction method using a face physiology model requires very complex computations,therefore,a gabor-based method was developed to synthesizerealistic facial expressions. In this paper, expression feature vectors are extracted by Gabor wavelet to synthesis the natural facial expression. To translate facial expressions,test results show that the method can successfully translate realistic facial expressions from one person to another.