中国科学院深圳先进技术研究院机构知识库(SIAT OpenIR): Deep Learning Face Representation from Predicting 10000 Classes
SIAT OpenIR  > 集成所
Deep Learning Face Representation from Predicting 10000 Classes
Yi Sun; Xiaogang Wang; Xiaoou Tang
2014
Conference NameComputer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on
Conference Place美国
AbstractThis paper proposes to learn a set of high-level feature representations through deep learning, referred to as Deep hidden IDentity features (DeepID), for face verification. We argue that DeepID can be effectively learned through challenging multi-class face identification tasks, whilst they can be generalized to other tasks (such as verification) and new identities unseen in the training set. Moreover, the generalization capability of DeepID increases as more face classes are to be predicted at training. DeepID features are taken from the last hidden layer neuron activations of deep convolutional networks (ConvNets). When learned as classifiers to recognize about 10,000 face identities in the training set and configured to keep reducing the neuron numbers along the feature extraction hierarchy, these deep ConvNets gradually form compact identity-related features in the top layers with only a small number of hidden neurons. The proposed features are extracted from various face regions to form complementary and over-complete representations. Any state-of-the-art classifiers can be learned based on these high-level representations for face verification. 97.45% verification accuracy on LFW is achieved with only weakly aligned faces.
Department多媒体集成技术研究中心
Indexed ByEI
Language英语
Document Type会议论文
Identifierhttp://ir.siat.ac.cn/handle/172644/5499
Collection集成所
Affiliation2014
Recommended Citation
GB/T 7714
Yi Sun,Xiaogang Wang,Xiaoou Tang. Deep Learning Face Representation from Predicting 10000 Classes[C],2014.
Files in This Item: Download All
File Name/Size DocType Version Access License
集成-多媒体2014018.pdf(2556KB) 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Yi Sun]'s Articles
[Xiaogang Wang]'s Articles
[Xiaoou Tang]'s Articles
Baidu academic
Similar articles in Baidu academic
[Yi Sun]'s Articles
[Xiaogang Wang]'s Articles
[Xiaoou Tang]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Yi Sun]'s Articles
[Xiaogang Wang]'s Articles
[Xiaoou Tang]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 集成-多媒体2014018.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.