摘要
为了实现三维人脸识别算法对表情变化的鲁棒性,提出一种基于语义对齐的多区域模板融合三维人脸识别算法。首先,为了实现三维人脸在语义上的对齐,将所有三维人脸模型与预定义标准参考模型做稠密对齐。然后,根据人脸表情具有区域性的特点,为了不受限于区域划分的精准度,提出基于多区域模板的相似度预测方法。最后,采用多数投票法将多个分类器的预测结果融合得到最终识别结果。实验结果表明,在FRGC v2.0表情三维人脸数据库上所提算法可以达到98.69%的rank-1识别率,在含有遮挡变化的Bosphorus数据库上该算法达到84.36%的rank-1识别率。
In order to realize the robustness of three-Dimensional(3 D) face recognition algorithm to expression variations, a multi-region template fusion 3 D face recognition algorithm based on semantic alignment was proposed. Firstly, in order to guarantee the semantic alignment of 3 D faces, all the 3 D face models were densely aligned with a pre-defined standard reference 3 D face model. Then, considering the expressions were regional, to be robust to region division, a multi-region template based similarity prediction method was proposed. Finally, all the prediction results of multiple classifiers were fused by majority voting method. The experimental results show that, the proposed algorithm can achieve the rank-1 face recognition rate of 98.69% on FRGC(the Face Recognition Grand Challenge) v2.0 expression 3 D face database and rank-1 face recognition rate of 84.36% on Bosphorus database with occlusion change.
作者
桑高丽
闫超
朱蓉
SANG Gaoli;YAN Chao;ZHU Rong(College of Mathematics and Information Engineering,Jiaxing University,Jiaxing Zhejiang 314001,China;College of Computer Science,Sichuan University,Chengdu Sichuan 610064,China)
出处
《计算机应用》
CSCD
北大核心
2019年第6期1685-1689,共5页
journal of Computer Applications
基金
国家自然科学基金资助项目(61703183)
浙江省自然科学基金资助项目(LY15F020039,LQ18F020007,LQ18F020006)~~
关键词
表情变化
三维人脸识别
多区域模板
多数投票
expression variation
three-Dimensional(3D) face recognition
multi-region template
majority voting
