姿态和表情变化下的三维人脸标志点定位（英文）

发布时间：2018-04-18 15:30

本文选题：人脸标志点定位 + HK曲率分析　；参考：《控制理论与应用》2017年06期

【摘要】：三维人脸标志点定位在人脸识别、人脸跟踪、人脸建模、表情分析等方面具有非常重要的作用.然而,在姿态和表情变化很大的条件下进行标志点定位,这仍然是一个很具挑战性的课题.本文提出一种对姿态和表情不敏感的三维人脸标志点定位方法,利用HK曲率分析检测出候选标志点,根据对面部形状的先验知识,提出一种基于人脸几何结构的分类策略对候选标志点进一步细分,通过把候选标志点与人脸标志点模型进行匹配,实现标志点的精确定位.首先在CASIA数据集对该方法的标志点定位精度进行测试,然后在UND/FRGC v2.0数据集对该方法与其他方法进行比较.实验结果表明该方法在姿态和表情变化很大的条件下具有高精度和高鲁棒性.
[Abstract]:3D face marker location plays an important role in face recognition, face tracking, face modeling, expression analysis and so on.However, it is still a challenging task to locate landmarks under the condition of great changes in posture and expression.In this paper, a 3D face marker location method, which is insensitive to pose and expression, is proposed. The candidate marker points are detected by HK curvature analysis, and based on the prior knowledge of facial shape,A classification strategy based on geometric structure of human face is proposed to further subdivide candidate markers. By matching candidate markers with face marker models, the precise location of candidate markers is realized.Firstly, the accuracy of the method is tested in the CASIA dataset, and then compared with other methods in the UND/FRGC v2.0 dataset.The experimental results show that the method has high accuracy and robustness under the condition of great changes in posture and expression.
【作者单位】：广东工业大学自动化学院;华南师范大学软件学院;
【分类号】：TP391.41

【相似文献】