A Double-Deep Spatio-Angular Learning Framework for Light Field based Face Recognition

Alireza Sepas-Moghaddam, Mohammad Ahsanul Haque, Paulo Lobato Correia, Kamal Nasrollahi, Thomas B. Moeslund, Fernando Pereira

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Face recognition has attracted increasing attention due to its wide range of applications, but it is still challenging when facing large variations in the biometric data characteristics. Lenslet light field cameras have recently come into prominence to capture rich spatio-angular information, thus offering new possibilities for advanced biometric recognition systems. This paper proposes a double-deep spatio-angular learning framework for light field based face recognition, which is able to model both the intra-view/spatial and inter-view/angular information using two deep networks in sequence. This is a novel recognition framework that has never been proposed in the literature for face recognition or any other visual recognition task. The proposed double-deep learning framework includes a long short-term memory (LSTM) recurrent network whose inputs are VGG-Face descriptions, computed using a VGG- 16 convolutional neural network (CNN). The VGG-Face spatial descriptions are extracted from a selected set of 2D sub-aperture (SA) images rendered from the light field image, corresponding to different observation angles. A sequence of the VGG-Face spatial descriptions is then be analysed by the LSTM network. A comprehensive set of experiments has been conducted using the IST-EURECOM light field face database, addressing varied and challenging recognition tasks. Results show that the proposed framework achieves superior face recognition performance when compared to the state-of-the-art.
Original languageEnglish
JournalI E E E Transactions on Circuits and Systems for Video Technology
ISSN1051-8215
Publication statusAccepted/In press - 8 May 2019

Fingerprint

Face recognition
Biometrics
Cameras
Neural networks
Experiments
Long short-term memory

Keywords

  • Face Recognition
  • Lenslet Light Field Imaging
  • Spatio-Angular Information
  • Deep Learning
  • VGG-Face Descriptor
  • VGG-Very-Deep-16 CNN
  • Long Short-Term Memory Network

Cite this

@article{18da86205754469896b268df35174379,
title = "A Double-Deep Spatio-Angular Learning Framework for Light Field based Face Recognition",
abstract = "Face recognition has attracted increasing attention due to its wide range of applications, but it is still challenging when facing large variations in the biometric data characteristics. Lenslet light field cameras have recently come into prominence to capture rich spatio-angular information, thus offering new possibilities for advanced biometric recognition systems. This paper proposes a double-deep spatio-angular learning framework for light field based face recognition, which is able to model both the intra-view/spatial and inter-view/angular information using two deep networks in sequence. This is a novel recognition framework that has never been proposed in the literature for face recognition or any other visual recognition task. The proposed double-deep learning framework includes a long short-term memory (LSTM) recurrent network whose inputs are VGG-Face descriptions, computed using a VGG- 16 convolutional neural network (CNN). The VGG-Face spatial descriptions are extracted from a selected set of 2D sub-aperture (SA) images rendered from the light field image, corresponding to different observation angles. A sequence of the VGG-Face spatial descriptions is then be analysed by the LSTM network. A comprehensive set of experiments has been conducted using the IST-EURECOM light field face database, addressing varied and challenging recognition tasks. Results show that the proposed framework achieves superior face recognition performance when compared to the state-of-the-art.",
keywords = "Face Recognition, Lenslet Light Field Imaging, Spatio-Angular Information, Deep Learning, VGG-Face Descriptor, VGG-Very-Deep-16 CNN, Long Short-Term Memory Network",
author = "Alireza Sepas-Moghaddam and Haque, {Mohammad Ahsanul} and Correia, {Paulo Lobato} and Kamal Nasrollahi and Moeslund, {Thomas B.} and Fernando Pereira",
year = "2019",
month = "5",
day = "8",
language = "English",
journal = "I E E E Transactions on Circuits and Systems for Video Technology",
issn = "1051-8215",
publisher = "IEEE",

}

A Double-Deep Spatio-Angular Learning Framework for Light Field based Face Recognition. / Sepas-Moghaddam, Alireza ; Haque, Mohammad Ahsanul; Correia, Paulo Lobato; Nasrollahi, Kamal; Moeslund, Thomas B.; Pereira, Fernando.

In: I E E E Transactions on Circuits and Systems for Video Technology, 08.05.2019.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - A Double-Deep Spatio-Angular Learning Framework for Light Field based Face Recognition

AU - Sepas-Moghaddam, Alireza

AU - Haque, Mohammad Ahsanul

AU - Correia, Paulo Lobato

AU - Nasrollahi, Kamal

AU - Moeslund, Thomas B.

AU - Pereira, Fernando

PY - 2019/5/8

Y1 - 2019/5/8

N2 - Face recognition has attracted increasing attention due to its wide range of applications, but it is still challenging when facing large variations in the biometric data characteristics. Lenslet light field cameras have recently come into prominence to capture rich spatio-angular information, thus offering new possibilities for advanced biometric recognition systems. This paper proposes a double-deep spatio-angular learning framework for light field based face recognition, which is able to model both the intra-view/spatial and inter-view/angular information using two deep networks in sequence. This is a novel recognition framework that has never been proposed in the literature for face recognition or any other visual recognition task. The proposed double-deep learning framework includes a long short-term memory (LSTM) recurrent network whose inputs are VGG-Face descriptions, computed using a VGG- 16 convolutional neural network (CNN). The VGG-Face spatial descriptions are extracted from a selected set of 2D sub-aperture (SA) images rendered from the light field image, corresponding to different observation angles. A sequence of the VGG-Face spatial descriptions is then be analysed by the LSTM network. A comprehensive set of experiments has been conducted using the IST-EURECOM light field face database, addressing varied and challenging recognition tasks. Results show that the proposed framework achieves superior face recognition performance when compared to the state-of-the-art.

AB - Face recognition has attracted increasing attention due to its wide range of applications, but it is still challenging when facing large variations in the biometric data characteristics. Lenslet light field cameras have recently come into prominence to capture rich spatio-angular information, thus offering new possibilities for advanced biometric recognition systems. This paper proposes a double-deep spatio-angular learning framework for light field based face recognition, which is able to model both the intra-view/spatial and inter-view/angular information using two deep networks in sequence. This is a novel recognition framework that has never been proposed in the literature for face recognition or any other visual recognition task. The proposed double-deep learning framework includes a long short-term memory (LSTM) recurrent network whose inputs are VGG-Face descriptions, computed using a VGG- 16 convolutional neural network (CNN). The VGG-Face spatial descriptions are extracted from a selected set of 2D sub-aperture (SA) images rendered from the light field image, corresponding to different observation angles. A sequence of the VGG-Face spatial descriptions is then be analysed by the LSTM network. A comprehensive set of experiments has been conducted using the IST-EURECOM light field face database, addressing varied and challenging recognition tasks. Results show that the proposed framework achieves superior face recognition performance when compared to the state-of-the-art.

KW - Face Recognition

KW - Lenslet Light Field Imaging

KW - Spatio-Angular Information

KW - Deep Learning

KW - VGG-Face Descriptor

KW - VGG-Very-Deep-16 CNN

KW - Long Short-Term Memory Network

M3 - Journal article

JO - I E E E Transactions on Circuits and Systems for Video Technology

JF - I E E E Transactions on Circuits and Systems for Video Technology

SN - 1051-8215

ER -