基于改进Morphing算法的人脸动画生成算法  

Facial animation generation algorithm based on improved Morphing algorithm

作　　者：黄晓瑜[1] HUANG Xiaoyu(School of Information Technology,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]桂林电子科技大学信息科技学院,广西桂林541004

出　　处：《现代电子技术》2020年第22期82-85,共4页Modern Electronics Technique

基　　金：广西创新驱动发展专项(科技重大专项)项目(桂科AA17204006)。

摘　　要：针对人脸渐变动画真实性和过渡效果难以提高的问题,文中基于图形渐变动画技术,提出一种改进的Morphing算法。该算法通过多次使用Adaboost算法以提高人脸特征定位的准确性和效率,同时又通过彩色和灰色图像的混合交叉渐变处理提高了人脸图像渐变效果的稳定性和真实性。经过测试与实验结果表明,该算法在未大幅提高计算量的同时,获得了真实性、稳定性和普适性较好的人脸渐变动画效果,且渐变过渡较为自然,为当前人脸动画生成的需求提供一种较为理想的技术解决方案。In allusion to the problem that it is difficult to improve the authenticity and transition effect of facial morphing animation,an improved Morphing algorithm is proposed based on graphics morphing animation technology.The accuracy and efficiency of facial feature location are improved by means of the Adaboost algorithm for more times,and the hybrid cross morphing processing of color and gray images is performed to improve the stability and authenticity of the morphing effect of the facial image.The testing and experimental results show that the algorithm not only does not greatly increase the amount of computation,but also achieves the effect of facial morphing animation with good authenticity,stability and universality,and more natural morphing transition,which provides an ideal technical solution for the current demand of facial animation generation.

关 键 词：人脸动画 Morphing算法 图像渐变 ADABOOST算法 人脸特征定位 混合交叉渐变 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP393[电子电信—信息与通信工程]