论文部分内容阅读
本文提出一种用于特定人逼真表情图像合成的新方法该方法利用二维网格描述脸部呈现不同表情时的几何变化，并使用主分量分析方法将该网格序列用参数化模型表示，从而极大地压缩了数据在重构表情图像时利用参考帧提供的场景纹理信息，结合网格参数模型提供的几何信息，使用图像卷曲与插值技术，生成逼真表情图像．目前使用该方法对人脸下半部表情图像分析及合成进行了初步的实验、实验结果表明了该种方法的有效性
In this paper, a new method for the synthesis of realistic human face images is proposed. This method uses two-dimensional grids to describe the geometric changes of faces with different facial expressions. The principal component analysis method is used to represent the grid sequence with parametric models. Which greatly compressed the scene texture information provided by the reference frame when reconstructing the emoticons, and combined with the geometric information provided by the grid parameter model, using the image curl and interpolation technology to generate realistic emoticon images. At present, this method is used to analyze and synthesize facial expression images in the lower half of the face. The experimental results show the effectiveness of this method