Learning Face Representation from Scratch
Learning Face Representation from Scratch
复制标题
DOI：
--
复制
发表时间：
2014-11
期刊：
影响因子：
--
通讯作者：
Dong Yi;Zhen Lei;Shengcai Liao;S. Li
中科院分区：
文献类型：
--
作者：
Dong Yi;Zhen Lei;Shengcai Liao;S. Li
在大数据和深度卷积神经网络（CNN）的推动下，人脸识别的性能正在变得与人类相当。使用私有大规模训练数据集，几个组在LFW上实现了非常高的性能，即，97%到99%。虽然CNN有许多开源实现，但没有一个大规模的人脸数据集是公开的。人脸识别领域的现状是数据比算法更重要。针对这一问题，本文提出了一种半自动的方法从互联网上采集人脸图像，并建立了一个包含约10，000个主题和500，000张图像的大规模数据集，称为CASIAWebFace。基于数据库，我们使用11层CNN来学习判别式表示，并在LFW和YTF上获得最先进的准确性。CASIAWebFace的发表将吸引更多的研究小组进入这一领域，加速人脸识别在野外的发展。
Pushing by big data and deep convolutional neural network (CNN), the performance of face recognition is becoming comparable to human. Using private large scale training datasets, several groups achieve very high performance on LFW, i.e., 97% to 99%. While there are many open source implementations of CNN, none of large scale face dataset is publicly available. The current situation in the field of face recognition is that data is more important than algorithm. To solve this problem, this paper proposes a semi-automatical way to collect face images from Internet and builds a large scale dataset containing about 10,000 subjects and 500,000 images, called CASIAWebFace. Based on the database, we use a 11-layer CNN to learn discriminative representation and obtain state-of-theart accuracy on LFW and YTF. The publication of CASIAWebFace will attract more research groups entering this field and accelerate the development of face recognition in the wild.