| 摘要: |
| 针对大姿态人脸对齐算法中的精确度低的问题，设计并实现了一种新的分层并行和多尺度Inception-Resnet网络来实现大姿态人脸对齐.首先，构建了一个四阶级联沙漏网络模型.该模型通过端到端的方式直接输入图像进行人脸对齐.其次，网络内部使用预先设定的参数进行采样和特征提取.最后，直接输出对应的人脸特征点提取图像以及同等人脸大小的二维坐标点绘制图，并将所提出的方法在AFLW2000-3D数据集上进行测试.实验结果表明，对于任意无约束的二维人脸图像，该方法的归一化平均误差为4.41%.与传统方法相比，该方法输出的正脸姿态图像视觉质量高、保真度更强. |
| 关键词: 人脸对齐 卷积神经网络 大姿态人脸 沙漏网络 端到端 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金（61741111） |
|
| Large-pose Face Alignment Based on Deep Learning |
|
JIANG Yue-Hui1, ZHANG Qian1, WANG Bin1, SHEN Hui-Zhong1, HUANG Ji-Feng1, YAN Tao2
|
|
1.School of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China;2.School of Information Engineering, Putian University, Putian 351100, China
|
| Abstract: |
| Aiming at the low accuracy of the large-pose face alignment algorithm, this paper designs and implements a new hierarchical parallel and multi-scale Inception-resnet network to achieve large-pose face alignment. Firstly, a four-class Hourglass network model is constructed. The model directly inputs images for face alignment in an end-to-end manner. Secondly, the network internally uses pre-set parameters for sampling and feature extraction. Finally, the corresponding face feature points are directly output. A two-dimensional coordinate point drawing of the image and the equivalent face size is extracted, and the proposed method is tested on the AFLW2000-3D data set. Experimental results show that the normalized average error of this method is 4.41% for any unconstrained two-dimensional face image. Compared with the traditional method, the positive face attitude image outputted in this paper has high visual quality and fidelity. |
| Key words: face alignment convolutional neural network large-pose face hourglass network end-to-end |
