Affordable Access

Access to the full text

Large-Scale Bisample Learning on ID Versus Spot Face Recognition

Authors
  • Zhu, Xiangyu1, 2
  • Liu, Hao1, 2
  • Lei, Zhen1, 2
  • Shi, Hailin1
  • Yang, Fan3
  • Yi, Dong4
  • Qi, Guojun5
  • Li, Stan Z.1, 2
  • 1 Chinese Academy of Sciences, Center for Biometrics and Security Research and National Laboratory of Pattern Recognition, Institute of Automation, Beijing, China , Beijing (China)
  • 2 University of Chinese Academy of Sciences, Beijing, China , Beijing (China)
  • 3 Beihang University, College of Software, Beijing, China , Beijing (China)
  • 4 Alibaba Group, DAMO Academy, Zhejiang, China , Zhejiang (China)
  • 5 HUAWEI Cloud, Boston, USA , Boston (United States)
Type
Published Article
Journal
International Journal of Computer Vision
Publisher
Springer-Verlag
Publication Date
Feb 16, 2019
Volume
127
Issue
6-7
Pages
684–700
Identifiers
DOI: 10.1007/s11263-019-01162-8
Source
Springer Nature
Keywords
License
Yellow

Abstract

In real-world face recognition applications, there is a tremendous amount of data with two images for each person. One is an ID photo for face enrollment, and the other is a probe photo captured on spot. Most existing methods are designed for training data with limited breadth (a relatively small number of classes) and sufficient depth (many samples for each class). They would meet great challenges on ID versus Spot (IvS) data, including the under-represented intra-class variations and an excessive demand on computing devices. In this paper, we propose a deep learning based large-scale bisample learning (LBL) method for IvS face recognition. To tackle the bisample problem with only two samples for each class, a classification–verification–classification training strategy is proposed to progressively enhance the IvS performance. Besides, a dominant prototype softmax is incorporated to make the deep learning scalable on large-scale classes. We conduct LBL on a IvS face dataset with more than two million identities. Experimental results show the proposed method achieves superior performance to previous ones, validating the effectiveness of LBL on IvS face recognition.

Report this publication

Statistics

Seen <100 times