Cargando…
Adaptive Spatial Transformation Networks for Periocular Recognition
Periocular recognition has emerged as a particularly valuable biometric identification method in challenging scenarios, such as partially occluded faces due to COVID-19 protective masks masks, in which face recognition might not be applicable. This work presents a periocular recognition framework ba...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10007286/ https://www.ncbi.nlm.nih.gov/pubmed/36904666 http://dx.doi.org/10.3390/s23052456 |
_version_ | 1784905482302914560 |
---|---|
author | Borza, Diana Laura Yaghoubi, Ehsan Frintrop, Simone Proença, Hugo |
author_facet | Borza, Diana Laura Yaghoubi, Ehsan Frintrop, Simone Proença, Hugo |
author_sort | Borza, Diana Laura |
collection | PubMed |
description | Periocular recognition has emerged as a particularly valuable biometric identification method in challenging scenarios, such as partially occluded faces due to COVID-19 protective masks masks, in which face recognition might not be applicable. This work presents a periocular recognition framework based on deep learning, which automatically localises and analyses the most important areas in the periocular region. The main idea is to derive several parallel local branches from a neural network architecture, which in a semi-supervised manner learn the most discriminative areas in the feature map and solve the identification problem solely upon the corresponding cues. Here, each local branch learns a transformation matrix that allows for basic geometrical transformations (cropping and scaling), which is used to select a region of interest in the feature map, further analysed by a set of shared convolutional layers. Finally, the information extracted by the local branches and the main global branch are fused together for recognition. The experiments carried out on the challenging UBIRIS-v2 benchmark show that by integrating the proposed framework with various ResNet architectures, we consistently obtain an improvement in mAP of more than 4% over the “vanilla” architecture. In addition, extensive ablation studies were performed to better understand the behavior of the network and how the spatial transformation and the local branches influence the overall performance of the model. The proposed method can be easily adapted to other computer vision problems, which is also regarded as one of its strengths. |
format | Online Article Text |
id | pubmed-10007286 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-100072862023-03-12 Adaptive Spatial Transformation Networks for Periocular Recognition Borza, Diana Laura Yaghoubi, Ehsan Frintrop, Simone Proença, Hugo Sensors (Basel) Article Periocular recognition has emerged as a particularly valuable biometric identification method in challenging scenarios, such as partially occluded faces due to COVID-19 protective masks masks, in which face recognition might not be applicable. This work presents a periocular recognition framework based on deep learning, which automatically localises and analyses the most important areas in the periocular region. The main idea is to derive several parallel local branches from a neural network architecture, which in a semi-supervised manner learn the most discriminative areas in the feature map and solve the identification problem solely upon the corresponding cues. Here, each local branch learns a transformation matrix that allows for basic geometrical transformations (cropping and scaling), which is used to select a region of interest in the feature map, further analysed by a set of shared convolutional layers. Finally, the information extracted by the local branches and the main global branch are fused together for recognition. The experiments carried out on the challenging UBIRIS-v2 benchmark show that by integrating the proposed framework with various ResNet architectures, we consistently obtain an improvement in mAP of more than 4% over the “vanilla” architecture. In addition, extensive ablation studies were performed to better understand the behavior of the network and how the spatial transformation and the local branches influence the overall performance of the model. The proposed method can be easily adapted to other computer vision problems, which is also regarded as one of its strengths. MDPI 2023-02-23 /pmc/articles/PMC10007286/ /pubmed/36904666 http://dx.doi.org/10.3390/s23052456 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Borza, Diana Laura Yaghoubi, Ehsan Frintrop, Simone Proença, Hugo Adaptive Spatial Transformation Networks for Periocular Recognition |
title | Adaptive Spatial Transformation Networks for Periocular Recognition |
title_full | Adaptive Spatial Transformation Networks for Periocular Recognition |
title_fullStr | Adaptive Spatial Transformation Networks for Periocular Recognition |
title_full_unstemmed | Adaptive Spatial Transformation Networks for Periocular Recognition |
title_short | Adaptive Spatial Transformation Networks for Periocular Recognition |
title_sort | adaptive spatial transformation networks for periocular recognition |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10007286/ https://www.ncbi.nlm.nih.gov/pubmed/36904666 http://dx.doi.org/10.3390/s23052456 |
work_keys_str_mv | AT borzadianalaura adaptivespatialtransformationnetworksforperiocularrecognition AT yaghoubiehsan adaptivespatialtransformationnetworksforperiocularrecognition AT frintropsimone adaptivespatialtransformationnetworksforperiocularrecognition AT proencahugo adaptivespatialtransformationnetworksforperiocularrecognition |