一种改进的遗传算法的三维人脸建模方法  

A method of improved GA for 3D face modeling

作　　者：孔素娟[1] 王成儒[1] 赵敏岑[1] 

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004

出　　处：《传感器与微系统》2011年第2期51-53,63,共4页Transducer and Microsystem Technologies

摘　　要：采用网格重采样算法提取人脸特征实现了真实人脸与原型三维人脸之间的精确对应,在此基础上,提出了基于改进的方向自学习遗传算法的模型匹配算法,该算法在局部搜索中引入方向信息,提高了算法的收敛速度和重建精度,实验结果表明:新算法可有效实现真实人脸与原型三维人脸之间的精确对应,减少了优化参数,提高了模型重建精度和速度,缩短了模型匹配的时间。Mesh re-sampling algorithm is used to exact features and realize exact correspondence between the real 3D face and the 3D model. A model matching method based on direction genetic algorithm is proposed. In the algorithm, the directional information is introduced in local search process. Convergent speed and precision of the algorithm are improved. The experimental results show that this novel algorithm effectively realize precise alignment to the prototypes, reduces optimization parameters, improves the precision and efficiency of model matching and shorten the matching time.

关 键 词：形变模型 网格重采样 遗传算法 方向自学习 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]