Biologically Inspired Tensor Features

Yang Mu, Dacheng Tao, Xuelong Li, Fionn Murtagh

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

According to the research results reported in the past decades, it is well acknowledged that face recognition is not a trivial task. With the development of electronic devices, we are gradually revealing the secret of object recognition in the primate's visual cortex. Therefore, it is time to reconsider face recognition by using biologically inspired features. In this paper, we represent face images by utilizing the C1 units, which correspond to complex cells in the visual cortex, and pool over S1 units by using a maximum operation to reserve only the maximum response of each local area of S1 units. The new representation is termed C1 Face. Because C1 Face is naturally a third-order tensor (or a three dimensional array), we propose three-way discriminative locality alignment (TWDLA), an extension of the discriminative locality alignment, which is a top-level discriminate manifold learning-based subspace learning algorithm. TWDLA has the following advantages: (1) it takes third-order tensors as input directly so the structure information can be well preserved; (2) it models the local geometry over every modality of the input tensors so the spatial relations of input tensors within a class can be preserved; (3) it maximizes the margin between a tensor and tensors from other classes over each modality so it performs well for recognition tasks and (4) it has no under sampling problem. Extensive experiments on YALE and FERET datasets show (1) the proposed C1Face representation can better represent face images than raw pixels and (2) TWDLA can duly preserve both the local geometry and the discriminative information over every modality for recognition.

Original languageEnglish
Article number327
Pages (from-to)327-341
Number of pages15
JournalCognitive Computation
Volume1
Issue number4
DOIs
Publication statusPublished - 1 Dec 2009
Externally publishedYes

Fingerprint

Tensors
Visual Cortex
Learning
Face recognition
Primates
Geometry
Object recognition
Equipment and Supplies
Learning algorithms
Pixels
Research
Sampling
Recognition (Psychology)
Facial Recognition
Experiments

Cite this

Mu, Yang ; Tao, Dacheng ; Li, Xuelong ; Murtagh, Fionn. / Biologically Inspired Tensor Features. In: Cognitive Computation. 2009 ; Vol. 1, No. 4. pp. 327-341.
@article{b1648b786dca483b8645f7933f1fd08c,
title = "Biologically Inspired Tensor Features",
abstract = "According to the research results reported in the past decades, it is well acknowledged that face recognition is not a trivial task. With the development of electronic devices, we are gradually revealing the secret of object recognition in the primate's visual cortex. Therefore, it is time to reconsider face recognition by using biologically inspired features. In this paper, we represent face images by utilizing the C1 units, which correspond to complex cells in the visual cortex, and pool over S1 units by using a maximum operation to reserve only the maximum response of each local area of S1 units. The new representation is termed C1 Face. Because C1 Face is naturally a third-order tensor (or a three dimensional array), we propose three-way discriminative locality alignment (TWDLA), an extension of the discriminative locality alignment, which is a top-level discriminate manifold learning-based subspace learning algorithm. TWDLA has the following advantages: (1) it takes third-order tensors as input directly so the structure information can be well preserved; (2) it models the local geometry over every modality of the input tensors so the spatial relations of input tensors within a class can be preserved; (3) it maximizes the margin between a tensor and tensors from other classes over each modality so it performs well for recognition tasks and (4) it has no under sampling problem. Extensive experiments on YALE and FERET datasets show (1) the proposed C1Face representation can better represent face images than raw pixels and (2) TWDLA can duly preserve both the local geometry and the discriminative information over every modality for recognition.",
keywords = "Biologically inspired features, C1 units, Discriminative locality alignment, Face recognition, Manifold learning",
author = "Yang Mu and Dacheng Tao and Xuelong Li and Fionn Murtagh",
year = "2009",
month = "12",
day = "1",
doi = "10.1007/s12559-009-9028-5",
language = "English",
volume = "1",
pages = "327--341",
journal = "Cognitive Computation",
issn = "1866-9956",
publisher = "Springer New York",
number = "4",

}

Biologically Inspired Tensor Features. / Mu, Yang; Tao, Dacheng; Li, Xuelong; Murtagh, Fionn.

In: Cognitive Computation, Vol. 1, No. 4, 327, 01.12.2009, p. 327-341.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Biologically Inspired Tensor Features

AU - Mu, Yang

AU - Tao, Dacheng

AU - Li, Xuelong

AU - Murtagh, Fionn

PY - 2009/12/1

Y1 - 2009/12/1

N2 - According to the research results reported in the past decades, it is well acknowledged that face recognition is not a trivial task. With the development of electronic devices, we are gradually revealing the secret of object recognition in the primate's visual cortex. Therefore, it is time to reconsider face recognition by using biologically inspired features. In this paper, we represent face images by utilizing the C1 units, which correspond to complex cells in the visual cortex, and pool over S1 units by using a maximum operation to reserve only the maximum response of each local area of S1 units. The new representation is termed C1 Face. Because C1 Face is naturally a third-order tensor (or a three dimensional array), we propose three-way discriminative locality alignment (TWDLA), an extension of the discriminative locality alignment, which is a top-level discriminate manifold learning-based subspace learning algorithm. TWDLA has the following advantages: (1) it takes third-order tensors as input directly so the structure information can be well preserved; (2) it models the local geometry over every modality of the input tensors so the spatial relations of input tensors within a class can be preserved; (3) it maximizes the margin between a tensor and tensors from other classes over each modality so it performs well for recognition tasks and (4) it has no under sampling problem. Extensive experiments on YALE and FERET datasets show (1) the proposed C1Face representation can better represent face images than raw pixels and (2) TWDLA can duly preserve both the local geometry and the discriminative information over every modality for recognition.

AB - According to the research results reported in the past decades, it is well acknowledged that face recognition is not a trivial task. With the development of electronic devices, we are gradually revealing the secret of object recognition in the primate's visual cortex. Therefore, it is time to reconsider face recognition by using biologically inspired features. In this paper, we represent face images by utilizing the C1 units, which correspond to complex cells in the visual cortex, and pool over S1 units by using a maximum operation to reserve only the maximum response of each local area of S1 units. The new representation is termed C1 Face. Because C1 Face is naturally a third-order tensor (or a three dimensional array), we propose three-way discriminative locality alignment (TWDLA), an extension of the discriminative locality alignment, which is a top-level discriminate manifold learning-based subspace learning algorithm. TWDLA has the following advantages: (1) it takes third-order tensors as input directly so the structure information can be well preserved; (2) it models the local geometry over every modality of the input tensors so the spatial relations of input tensors within a class can be preserved; (3) it maximizes the margin between a tensor and tensors from other classes over each modality so it performs well for recognition tasks and (4) it has no under sampling problem. Extensive experiments on YALE and FERET datasets show (1) the proposed C1Face representation can better represent face images than raw pixels and (2) TWDLA can duly preserve both the local geometry and the discriminative information over every modality for recognition.

KW - Biologically inspired features

KW - C1 units

KW - Discriminative locality alignment

KW - Face recognition

KW - Manifold learning

UR - http://www.scopus.com/inward/record.url?scp=78649780115&partnerID=8YFLogxK

U2 - 10.1007/s12559-009-9028-5

DO - 10.1007/s12559-009-9028-5

M3 - Article

VL - 1

SP - 327

EP - 341

JO - Cognitive Computation

JF - Cognitive Computation

SN - 1866-9956

IS - 4

M1 - 327

ER -