DCTNet

A simple learning-free approach for face recognition

Cong Jie Ng, Beng Jin Teoh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

PCANet was proposed as a lightweight deep learning network that mainly leverages Principal Component Analysis (PCA) to learn multistage filter banks followed by binarization and block-wise histograming. PCANet was shown worked surprisingly well in various image classification tasks. However, PCANet is data-dependence hence inflexible. In this paper, we proposed a data-independence network, dubbed DCTNet for face recognition in which we adopt Discrete Cosine Transform (DCT) as filter banks in place of PCA. This is motivated by the fact that 2D DCT basis is indeed a good approximation for high ranked eigenvectors of PCA. Both 2D DCT and PCA resemble a kind of modulated sine-wave patterns, which can be perceived as a bandpass filter bank. DCTNet is free from learning as 2D DCT bases can be computed in advance. Besides that, we also proposed an effective method to regulate the block-wise histogram feature vector of DCTNet for robustness. It is shown to provide surprising performance boost when the probe image is considerably different in appearance from the gallery image. We evaluate the performance of DCTNet extensively on a number of benchmark face databases and being able to achieve on par with or often better accuracy performance than PCANet.

Original languageEnglish
Title of host publication2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages761-768
Number of pages8
ISBN (Electronic)9789881476807
DOIs
Publication statusPublished - 2016 Feb 19
Event2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015 - Hong Kong, Hong Kong
Duration: 2015 Dec 162015 Dec 19

Publication series

Name2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015

Other

Other2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015
CountryHong Kong
CityHong Kong
Period15/12/1615/12/19

Fingerprint

Discrete Cosine Transform
Discrete cosine transforms
Face recognition
Face Recognition
Principal component analysis
Principal Component Analysis
Filter Banks
Filter banks
Binarization
Data Dependence
Bandpass Filter
Image classification
Image Classification
Bandpass filters
Feature Vector
Eigenvalues and eigenfunctions
Leverage
Histogram
Eigenvector
Probe

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Modelling and Simulation
  • Signal Processing

Cite this

Ng, C. J., & Teoh, B. J. (2016). DCTNet: A simple learning-free approach for face recognition. In 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015 (pp. 761-768). [7415375] (2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/APSIPA.2015.7415375
Ng, Cong Jie ; Teoh, Beng Jin. / DCTNet : A simple learning-free approach for face recognition. 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 761-768 (2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015).
@inproceedings{c70d2011938044ec80f5b64b8d67f4af,
title = "DCTNet: A simple learning-free approach for face recognition",
abstract = "PCANet was proposed as a lightweight deep learning network that mainly leverages Principal Component Analysis (PCA) to learn multistage filter banks followed by binarization and block-wise histograming. PCANet was shown worked surprisingly well in various image classification tasks. However, PCANet is data-dependence hence inflexible. In this paper, we proposed a data-independence network, dubbed DCTNet for face recognition in which we adopt Discrete Cosine Transform (DCT) as filter banks in place of PCA. This is motivated by the fact that 2D DCT basis is indeed a good approximation for high ranked eigenvectors of PCA. Both 2D DCT and PCA resemble a kind of modulated sine-wave patterns, which can be perceived as a bandpass filter bank. DCTNet is free from learning as 2D DCT bases can be computed in advance. Besides that, we also proposed an effective method to regulate the block-wise histogram feature vector of DCTNet for robustness. It is shown to provide surprising performance boost when the probe image is considerably different in appearance from the gallery image. We evaluate the performance of DCTNet extensively on a number of benchmark face databases and being able to achieve on par with or often better accuracy performance than PCANet.",
author = "Ng, {Cong Jie} and Teoh, {Beng Jin}",
year = "2016",
month = "2",
day = "19",
doi = "10.1109/APSIPA.2015.7415375",
language = "English",
series = "2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "761--768",
booktitle = "2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015",
address = "United States",

}

Ng, CJ & Teoh, BJ 2016, DCTNet: A simple learning-free approach for face recognition. in 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015., 7415375, 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015, Institute of Electrical and Electronics Engineers Inc., pp. 761-768, 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015, Hong Kong, Hong Kong, 15/12/16. https://doi.org/10.1109/APSIPA.2015.7415375

DCTNet : A simple learning-free approach for face recognition. / Ng, Cong Jie; Teoh, Beng Jin.

2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 761-768 7415375 (2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - DCTNet

T2 - A simple learning-free approach for face recognition

AU - Ng, Cong Jie

AU - Teoh, Beng Jin

PY - 2016/2/19

Y1 - 2016/2/19

N2 - PCANet was proposed as a lightweight deep learning network that mainly leverages Principal Component Analysis (PCA) to learn multistage filter banks followed by binarization and block-wise histograming. PCANet was shown worked surprisingly well in various image classification tasks. However, PCANet is data-dependence hence inflexible. In this paper, we proposed a data-independence network, dubbed DCTNet for face recognition in which we adopt Discrete Cosine Transform (DCT) as filter banks in place of PCA. This is motivated by the fact that 2D DCT basis is indeed a good approximation for high ranked eigenvectors of PCA. Both 2D DCT and PCA resemble a kind of modulated sine-wave patterns, which can be perceived as a bandpass filter bank. DCTNet is free from learning as 2D DCT bases can be computed in advance. Besides that, we also proposed an effective method to regulate the block-wise histogram feature vector of DCTNet for robustness. It is shown to provide surprising performance boost when the probe image is considerably different in appearance from the gallery image. We evaluate the performance of DCTNet extensively on a number of benchmark face databases and being able to achieve on par with or often better accuracy performance than PCANet.

AB - PCANet was proposed as a lightweight deep learning network that mainly leverages Principal Component Analysis (PCA) to learn multistage filter banks followed by binarization and block-wise histograming. PCANet was shown worked surprisingly well in various image classification tasks. However, PCANet is data-dependence hence inflexible. In this paper, we proposed a data-independence network, dubbed DCTNet for face recognition in which we adopt Discrete Cosine Transform (DCT) as filter banks in place of PCA. This is motivated by the fact that 2D DCT basis is indeed a good approximation for high ranked eigenvectors of PCA. Both 2D DCT and PCA resemble a kind of modulated sine-wave patterns, which can be perceived as a bandpass filter bank. DCTNet is free from learning as 2D DCT bases can be computed in advance. Besides that, we also proposed an effective method to regulate the block-wise histogram feature vector of DCTNet for robustness. It is shown to provide surprising performance boost when the probe image is considerably different in appearance from the gallery image. We evaluate the performance of DCTNet extensively on a number of benchmark face databases and being able to achieve on par with or often better accuracy performance than PCANet.

UR - http://www.scopus.com/inward/record.url?scp=84986224301&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986224301&partnerID=8YFLogxK

U2 - 10.1109/APSIPA.2015.7415375

DO - 10.1109/APSIPA.2015.7415375

M3 - Conference contribution

T3 - 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015

SP - 761

EP - 768

BT - 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Ng CJ, Teoh BJ. DCTNet: A simple learning-free approach for face recognition. In 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 761-768. 7415375. (2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2015). https://doi.org/10.1109/APSIPA.2015.7415375