论文部分内容阅读
本文提出了一种基于深度卷积神经网络的跨年龄人脸识别方法。此方法与传统人脸识别技术不同,它不需要人工进行人脸特征的设计与提取,能够通过在大量带有标签的人脸照片组中通过有监督学习自动抽取人脸特征并且能提取到对分类问题更有用的人脸特征表达和图像模式的隐性规律。我们希望此跨年领人脸识别算法与人类大脑类似,能够自动识别一对跨年龄阶段的照片是否为同一个人的照片。
This paper presents a cross-age face recognition method based on deep convolution neural network. This method is different from the traditional face recognition technology, it does not need to manually design and extract facial features, face images through a large number of tagged groups by supervised learning automatically extract facial features and can be extracted to Classification problem more useful face expression and image model of the recessive law. We hope this New Year’s Eve face recognition algorithm is similar to the human brain and automatically recognizes whether a photo of the same individual is across the ages.