Knowledge-based tensor subspace analysis ...
Type de document :
Compte-rendu et recension critique d'ouvrage
Titre :
Knowledge-based tensor subspace analysis system for kinship verification
Auteur(s) :
Serraoui, I. [Auteur]
University of Biskra Mohamed Khider
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Laiadi, O. [Auteur]
University of Biskra Mohamed Khider
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Ouamane, A. [Auteur]
University of Biskra Mohamed Khider
Dornaika, F. [Auteur]
Universidad del Pais Vasco / Euskal Herriko Unibertsitatea [Espagne] [UPV/EHU]
Henan Polytechnic University
Ikerbasque - Basque Foundation for Science
Tahleb Ahmed, Abdelmalik [Auteur]
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
COMmunications NUMériques - IEMN [COMNUM - IEMN]
University of Biskra Mohamed Khider
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Laiadi, O. [Auteur]
University of Biskra Mohamed Khider
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Ouamane, A. [Auteur]
University of Biskra Mohamed Khider
Dornaika, F. [Auteur]
Universidad del Pais Vasco / Euskal Herriko Unibertsitatea [Espagne] [UPV/EHU]
Henan Polytechnic University
Ikerbasque - Basque Foundation for Science
Tahleb Ahmed, Abdelmalik [Auteur]
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
COMmunications NUMériques - IEMN [COMNUM - IEMN]
Titre de la revue :
Neural Networks
Pagination :
222-237
Éditeur :
Elsevier
Date de publication :
2022-07
ISSN :
0893-6080
Mot(s)-clé(s) en anglais :
Kinship verification
Knowledge-based tensor subspace analysis
Convolutional neural networks
Multi-view deep features
Metric learning
Facial images analysis
Knowledge-based tensor subspace analysis
Convolutional neural networks
Multi-view deep features
Metric learning
Facial images analysis
Discipline(s) HAL :
Sciences de l'ingénieur [physics]
Résumé en anglais : [en]
Most existing automatic kinship verification methods focus on learning the optimal distance metrics between family members. However, learning facial features and kinship features simultaneously may cause the proposed models ...
Lire la suite >Most existing automatic kinship verification methods focus on learning the optimal distance metrics between family members. However, learning facial features and kinship features simultaneously may cause the proposed models to be too weak. In this work, we explore the possibility of bridging this gap by developing knowledge-based tensor models based on pre-trained multi-view models. We propose an effective knowledge-based tensor similarity extraction framework for automatic facial kinship verification using four pre-trained networks (i.e., VGG-Face, VGG-F, VGG-M, and VGG-S). Therefore, knowledge-based deep face and general features (such as identity, age, gender, ethnicity, expression, lighting, pose, contour, edges, corners, shape, etc.) were successfully fused by our tensor design to understand the kinship cue. Multiple effective representations are learned for kinship verification statements (children and parents) using a margin maximization learning scheme based on Tensor Cross-view Quadratic Exponential Discriminant Analysis. Through the exponential learning process, the large gap between distributions of the same family can be reduced to the maximum, while the small gap between distributions of different families is simultaneously increased. The WCCN metric successfully reduces the intra-class variability problem caused by deep features. The explanation of black-box models and the problems of ubiquitous face recognition are considered in our system. The extensive experiments on four challenging datasets show that our system performs very well compared to state-of-the-art approaches.Lire moins >
Lire la suite >Most existing automatic kinship verification methods focus on learning the optimal distance metrics between family members. However, learning facial features and kinship features simultaneously may cause the proposed models to be too weak. In this work, we explore the possibility of bridging this gap by developing knowledge-based tensor models based on pre-trained multi-view models. We propose an effective knowledge-based tensor similarity extraction framework for automatic facial kinship verification using four pre-trained networks (i.e., VGG-Face, VGG-F, VGG-M, and VGG-S). Therefore, knowledge-based deep face and general features (such as identity, age, gender, ethnicity, expression, lighting, pose, contour, edges, corners, shape, etc.) were successfully fused by our tensor design to understand the kinship cue. Multiple effective representations are learned for kinship verification statements (children and parents) using a margin maximization learning scheme based on Tensor Cross-view Quadratic Exponential Discriminant Analysis. Through the exponential learning process, the large gap between distributions of the same family can be reduced to the maximum, while the small gap between distributions of different families is simultaneously increased. The WCCN metric successfully reduces the intra-class variability problem caused by deep features. The explanation of black-box models and the problems of ubiquitous face recognition are considered in our system. The extensive experiments on four challenging datasets show that our system performs very well compared to state-of-the-art approaches.Lire moins >
Langue :
Anglais
Vulgarisation :
Non
Source :