一种用于脸部特征检测的分层概率模型研究  

Research on hierarchical probability model for facial feature detection

作　　者：石贵民 余文森 SHI Guimin;YU Wensen(School of Mathematics and Computer, Wuyi University, Wuyishan, Fujian 354300, China;The Key Laboratory of Cognitive Computing and Intelligent Information Processing of Fujian Education Institutions,Wuyishan, Fujian 354300, China)

机构地区：[1]武夷学院数学与计算机学院,福建武夷山354300 [2]认知计算与智能信息处理福建省高校重点实验室,福建武夷山354300

出　　处：《计算机工程与应用》2017年第20期154-160,共7页Computer Engineering and Applications

基　　金：福建省自然科学基金(No.2015J01669);福建省高校专项基金(No.JK2015052);福建省教育厅重点科技项目(No.JA14309);福建省中青年教师教育科研基金(No.JB14099);福建省教育科学"十二五"规划2015年度课题(No.FJJKCG15-195);武夷学院校科研基金(No.XQ201306)

摘　　要：脸部特征检测问题是计算机视觉领域的研究热点。因为脸部外观和形态随着条件的变化而变化,因此面部特征检测比较复杂。针对现有脸部特征检测算法的不足,提出一种已知图像测量数据后能够推断出真实脸部特征位置的分层概率模型。针对每个脸部子部位的局部形态变化进行间接建模;通过搜索模型的最优结构和参数设置,在更高层次上学习脸部子部位、脸部表情和姿态间的联合关系。该模型综合利用了脸部子部位自下而上的形态约束以及脸部子部位间自上而下的关系约束来推断出脸部特征的真实位置。利用基准数据库进行了仿真实验。实验结果表明,该方法的检测性能要明显优于目前最新的人脸特征检测算法。Facial feature detection problem is a hot research topic in the field of computer vision.It is a nontrivial tasksince the appearance and shape of the face tend to change under different conditions.Aiming at the deficiency of the existingfacial feature detection algorithms,a hierarchical probabilistic model is proposed that can infer the true facial featurelocations given image measurements.Firstly,it implicitly models the local shape variation for each facial component.Secondly,it learns the joint relationship among facial components,the facial expression and the pose in the higher levelby searching the optimal structure and parameterizations of the model.Finally,the true facial feature locations are inferredthrough the bottom-up lower level shape constraints of facial components and the top-down constraint from the relationshipamong facial components.Experimental results on benchmark databases demonstrate that,the detection performanceof the proposed method is obviously better than that of the most recent face feature detection algorithm.

关 键 词：脸部特征检测 分层概率模型 局部形态 约束 检测误差 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]