RUS  ENG JOURNALS   PEOPLE   ORGANISATIONS   CONFERENCES   SEMINARS   VIDEO LIBRARY   PACKAGE AMSBIB
General information
Latest issue
Archive

Search papers
Search references

RSS
Latest issue
Current issues
Archive issues
What is RSS



Computer Optics:
Year:
Volume:
Issue:
Page:
Find






Personal entry:
Login:
Password:
Save password
Enter
Forgotten password?
Register


Computer Optics, 2017, Volume 41, Issue 2, Pages 254–265 (Mi co382)  

IMAGE PROCESSING, PATTERN RECOGNITION

Real-time face identification via CNN and boosted hashing forest

Yu. V. Vizilter, V. S. Gorbatcevich, A. V. Vorotnikov, N. A. Kostromov

State Research Institute of Aviation Systems (GosNIIAS), Moscow, Russia

Abstract: This paper presents a new approach to constructing a biometric template using a Convolutional Neural Network (CNN) with Hashing Forest. The approach consists of several steps: training a convolutional neural network, transforming it to a multiple convolution architecture, and finally learning the output hashing transform via a new Boosted Hashing Forest technique. This technique generalizes the Boosted SSC (Similarity Sensitive Coding) approach for hashing learning with joint optimization of face verification and identification. The proposed network via hashing forest is trained on the CASIA-WebFace dataset and evaluated on the LFW dataset. The result of coding the output of a single CNN is 97% on LFW. For Hamming embedding, the proposed approach enables a 200 bit (25 byte) code to be constructed with a 96.3% verification accuracy and a 2000-bit code with a 98.14% verification accuracy on LFW. The convolutional network with hashing forest with 2000x7-bit hashing trees achieves 93% rank-1 on LFW relative to the basic convolutional network's 89.9% rank-1. The proposed approach generates templates at the rate of 40+ fps with a GPU Core i7 and 120+ fps with a GPU GeForce GTX 650.

Keywords: convolutional neural networks, hashing, binary trees, Hamming distance, biometrics.

Funding Agency Grant Number
Russian Science Foundation 16-11-00082
The work was supported by a grant from the Russian Science Foundation (Project No. 16-11-00082).


DOI: https://doi.org/10.18287/2412-6179-2017-41-2-254-265

Full text: PDF file (1001 kB)
Full text: http://www.computeroptics.smr.ru/.../410214.html
References: PDF file   HTML file

Received: 23.11.2016
Accepted:16.03.2017

Citation: Yu. V. Vizilter, V. S. Gorbatcevich, A. V. Vorotnikov, N. A. Kostromov, “Real-time face identification via CNN and boosted hashing forest”, Computer Optics, 41:2 (2017), 254–265

Citation in format AMSBIB
\Bibitem{VizGorVor17}
\by Yu.~V.~Vizilter, V.~S.~Gorbatcevich, A.~V.~Vorotnikov, N.~A.~Kostromov
\paper Real-time face identification via CNN and boosted hashing forest
\jour Computer Optics
\yr 2017
\vol 41
\issue 2
\pages 254--265
\mathnet{http://mi.mathnet.ru/co382}
\crossref{https://doi.org/10.18287/2412-6179-2017-41-2-254-265}


Linking options:
  • http://mi.mathnet.ru/eng/co382
  • http://mi.mathnet.ru/eng/co/v41/i2/p254

    SHARE: VKontakte.ru FaceBook Twitter Mail.ru Livejournal Memori.ru


    Citing articles on Google Scholar: Russian citations, English citations
    Related articles on Google Scholar: Russian articles, English articles
  • Computer Optics
    Number of views:
    This page:204
    Full text:70
    References:20

     
    Contact us:
     Terms of Use  Registration  Logotypes © Steklov Mathematical Institute RAS, 2020