Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition
Over the years, the evolution of face recognition (FR) algorithms has been steep and accelerated by a myriad of factors. Motivated by the unexpected elements found in real-world scenarios, researchers have investigated and developed a number of methods for occluded face recognition (OFR). However, d...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2022-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9857925/ |
_version_ | 1811339561887858688 |
---|---|
author | Pedro C. Pedro Neto Joao Ribeiro Pinto Fadi Boutros Naser Damer Ana F. Sequeira Jaime S. Cardoso |
author_facet | Pedro C. Pedro Neto Joao Ribeiro Pinto Fadi Boutros Naser Damer Ana F. Sequeira Jaime S. Cardoso |
author_sort | Pedro C. Pedro Neto |
collection | DOAJ |
description | Over the years, the evolution of face recognition (FR) algorithms has been steep and accelerated by a myriad of factors. Motivated by the unexpected elements found in real-world scenarios, researchers have investigated and developed a number of methods for occluded face recognition (OFR). However, due to the SarS-Cov2 pandemic, masked face recognition (MFR) research branched from OFR and became a hot and urgent research challenge. Due to time and data constraints, these models followed different and novel approaches to handle lower face occlusions, i.e., face masks. Hence, this study aims to evaluate the different approaches followed for both MFR and OFR, find linked details about the two conceptually similar research directions and understand future directions for both topics. For this analysis, several occluded and face recognition algorithms from the literature are studied. First, they are evaluated in the task that they were trained on, but also on the other. These methods were picked accordingly to the novelty of their approach, proven state-of-the-art results, and publicly available source code. We present quantitative results on 4 occluded and 5 masked FR datasets, and a qualitative analysis of several MFR and OFR models on the Occ-LFW dataset. The analysis presented, sustain the interoperable deployability of MFR methods on OFR datasets, when the occlusions are of a reasonable size. Thus, solutions proposed for MFR can be effectively deployed for general OFR. |
first_indexed | 2024-04-13T18:28:38Z |
format | Article |
id | doaj.art-5cb8fefa506f4c1980843389a9c8cdc8 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-13T18:28:38Z |
publishDate | 2022-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-5cb8fefa506f4c1980843389a9c8cdc82022-12-22T02:35:10ZengIEEEIEEE Access2169-35362022-01-0110862228623310.1109/ACCESS.2022.31990149857925Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face RecognitionPedro C. Pedro Neto0https://orcid.org/0000-0003-1333-4889Joao Ribeiro Pinto1Fadi Boutros2https://orcid.org/0000-0003-4516-9128Naser Damer3https://orcid.org/0000-0001-7910-7895Ana F. Sequeira4https://orcid.org/0000-0002-6685-2033Jaime S. Cardoso5https://orcid.org/0000-0002-3760-2473Centre for Telecommunications and Multimedia, INESC TEC, Porto, PortugalCentre for Telecommunications and Multimedia, INESC TEC, Porto, PortugalFraunhofer Institute for Computer Graphics Research IGD, Darmstadt, GermanyFraunhofer Institute for Computer Graphics Research IGD, Darmstadt, GermanyCentre for Telecommunications and Multimedia, INESC TEC, Porto, PortugalCentre for Telecommunications and Multimedia, INESC TEC, Porto, PortugalOver the years, the evolution of face recognition (FR) algorithms has been steep and accelerated by a myriad of factors. Motivated by the unexpected elements found in real-world scenarios, researchers have investigated and developed a number of methods for occluded face recognition (OFR). However, due to the SarS-Cov2 pandemic, masked face recognition (MFR) research branched from OFR and became a hot and urgent research challenge. Due to time and data constraints, these models followed different and novel approaches to handle lower face occlusions, i.e., face masks. Hence, this study aims to evaluate the different approaches followed for both MFR and OFR, find linked details about the two conceptually similar research directions and understand future directions for both topics. For this analysis, several occluded and face recognition algorithms from the literature are studied. First, they are evaluated in the task that they were trained on, but also on the other. These methods were picked accordingly to the novelty of their approach, proven state-of-the-art results, and publicly available source code. We present quantitative results on 4 occluded and 5 masked FR datasets, and a qualitative analysis of several MFR and OFR models on the Occ-LFW dataset. The analysis presented, sustain the interoperable deployability of MFR methods on OFR datasets, when the occlusions are of a reasonable size. Thus, solutions proposed for MFR can be effectively deployed for general OFR.https://ieeexplore.ieee.org/document/9857925/Deep learningbiometricsoccluded face recognitionmasked face recognitionface biometricscomputer vision |
spellingShingle | Pedro C. Pedro Neto Joao Ribeiro Pinto Fadi Boutros Naser Damer Ana F. Sequeira Jaime S. Cardoso Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition IEEE Access Deep learning biometrics occluded face recognition masked face recognition face biometrics computer vision |
title | Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition |
title_full | Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition |
title_fullStr | Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition |
title_full_unstemmed | Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition |
title_short | Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition |
title_sort | beyond masks on the generalization of masked face recognition models to occluded face recognition |
topic | Deep learning biometrics occluded face recognition masked face recognition face biometrics computer vision |
url | https://ieeexplore.ieee.org/document/9857925/ |
work_keys_str_mv | AT pedrocpedroneto beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition AT joaoribeiropinto beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition AT fadiboutros beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition AT naserdamer beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition AT anafsequeira beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition AT jaimescardoso beyondmasksonthegeneralizationofmaskedfacerecognitionmodelstooccludedfacerecognition |