Kinship recognition from faces using deep learning with imbalanced data

Kinship verification from faces aims to determine whether two person share some family relationship based only on the visual facial patterns. This has attracted a significant interests among the scientific community due to its potential applications in social media mining and finding missing childre...

Full description

Saved in:
Bibliographic Details
Main Author: Hadid, Abdenour (author)
Other Authors: Othmani, Alice (author), Han, Duqing (author), Gao, Xin (author), Ye, Runpeng (author)
Published: 2022
Subjects:
Online Access:https://depot.sorbonne.ae/handle/20.500.12458/1321
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1857415064326242305
author Hadid, Abdenour
author2 Othmani, Alice
Han, Duqing
Gao, Xin
Ye, Runpeng
author2_role author
author
author
author
author_facet Hadid, Abdenour
Othmani, Alice
Han, Duqing
Gao, Xin
Ye, Runpeng
author_role author
dc.creator.none.fl_str_mv Hadid, Abdenour
Othmani, Alice
Han, Duqing
Gao, Xin
Ye, Runpeng
dc.date.none.fl_str_mv 2022-11-03T05:27:22Z
2022-11-03T05:27:22Z
2022
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv 10.1007/s11042-022-14058-6
1380-7501
1573-7721
https://depot.sorbonne.ae/handle/20.500.12458/1321
10.1007/s11042-022-14058-6
dc.language.none.fl_str_mv en
dc.relation.none.fl_str_mv Multimedia Tools and Applications
dc.subject.none.fl_str_mv Human-computer interaction
Kinship recognition
Deep visual learning
Deep learning
Biometrics
dc.title.none.fl_str_mv Kinship recognition from faces using deep learning with imbalanced data
dc.type.none.fl_str_mv Controlled Vocabulary for Resource Type Genres::text::periodical::journal::contribution to journal::journal article
description Kinship verification from faces aims to determine whether two person share some family relationship based only on the visual facial patterns. This has attracted a significant interests among the scientific community due to its potential applications in social media mining and finding missing children. In this work, We propose a novel pattern analysis technique for kinship verification based on a new deep learning-based approach. More specifically, given a pair of face images, we first use Resnet50 to extract deep features from each image. Then, feature distances between each pair of images are computed. Importantly, to overcome the problem of unbalanced data, One Hot Encoding for labels is utilised. The distances finally are fed to a deep neural networks to determine the kinship relation. Extensive experiments are conducted on FIW dataset containing 11 classes of kinship relationships. The experiments showed very promising results and pointed out the importance of balancing the training dataset. Moreover, our approach showed interesting ability of generalization. Results show that our approach performs better than all existing approaches on grandparents-grandchildren type of kinship. To support the principle of open and reproducible research, we are soon making our code publicly available to the research community: github.com/Steven-HDQ/Kinship-Recognition.
id sorbonner_2beba486fc32b13e43eb134bfcd2b6db
identifier_str_mv 10.1007/s11042-022-14058-6
1380-7501
1573-7721
language_invalid_str_mv en
network_acronym_str sorbonner
network_name_str Sorbonne University Abu Dhabi repository
oai_identifier_str oai:depot.sorbonne.ae:20.500.12458/1321
publishDate 2022
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Kinship recognition from faces using deep learning with imbalanced dataHadid, AbdenourOthmani, AliceHan, DuqingGao, XinYe, RunpengHuman-computer interactionKinship recognitionDeep visual learningDeep learningBiometricsKinship verification from faces aims to determine whether two person share some family relationship based only on the visual facial patterns. This has attracted a significant interests among the scientific community due to its potential applications in social media mining and finding missing children. In this work, We propose a novel pattern analysis technique for kinship verification based on a new deep learning-based approach. More specifically, given a pair of face images, we first use Resnet50 to extract deep features from each image. Then, feature distances between each pair of images are computed. Importantly, to overcome the problem of unbalanced data, One Hot Encoding for labels is utilised. The distances finally are fed to a deep neural networks to determine the kinship relation. Extensive experiments are conducted on FIW dataset containing 11 classes of kinship relationships. The experiments showed very promising results and pointed out the importance of balancing the training dataset. Moreover, our approach showed interesting ability of generalization. Results show that our approach performs better than all existing approaches on grandparents-grandchildren type of kinship. To support the principle of open and reproducible research, we are soon making our code publicly available to the research community: github.com/Steven-HDQ/Kinship-Recognition.2022-11-03T05:27:22Z2022-11-03T05:27:22Z2022Controlled Vocabulary for Resource Type Genres::text::periodical::journal::contribution to journal::journal articleapplication/pdf10.1007/s11042-022-14058-61380-75011573-7721https://depot.sorbonne.ae/handle/20.500.12458/132110.1007/s11042-022-14058-6enMultimedia Tools and Applicationsoai:depot.sorbonne.ae:20.500.12458/13212022-11-15T18:00:40Z
spellingShingle Kinship recognition from faces using deep learning with imbalanced data
Hadid, Abdenour
Human-computer interaction
Kinship recognition
Deep visual learning
Deep learning
Biometrics
title Kinship recognition from faces using deep learning with imbalanced data
title_full Kinship recognition from faces using deep learning with imbalanced data
title_fullStr Kinship recognition from faces using deep learning with imbalanced data
title_full_unstemmed Kinship recognition from faces using deep learning with imbalanced data
title_short Kinship recognition from faces using deep learning with imbalanced data
title_sort Kinship recognition from faces using deep learning with imbalanced data
topic Human-computer interaction
Kinship recognition
Deep visual learning
Deep learning
Biometrics
url https://depot.sorbonne.ae/handle/20.500.12458/1321