Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques
Face Anti-Spoofing (FAS) systems are entrusted with the task of determining whether the content of a facial image is genuine or fake. The performance and overall quality of those systems depend on the data they are fed during their development stage, meaning that high-resolution and diverse facial d...
- Autores:
-
Avalos López, Fernando Andrés
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2024
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/73851
- Acceso en línea:
- https://hdl.handle.net/1992/73851
- Palabra clave:
- Face anti-spoofing
Biometric privacy-enhancing techniques
Fine-tuning
Performance analysis
Ingeniería
- Rights
- embargoedAccess
- License
- Attribution 4.0 International
| id |
UNIANDES2_0590069c481bfecb0e9afd34ae989dff |
|---|---|
| oai_identifier_str |
oai:repositorio.uniandes.edu.co:1992/73851 |
| network_acronym_str |
UNIANDES2 |
| network_name_str |
Séneca: repositorio Uniandes |
| repository_id_str |
|
| dc.title.eng.fl_str_mv |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| title |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| spellingShingle |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques Face anti-spoofing Biometric privacy-enhancing techniques Fine-tuning Performance analysis Ingeniería |
| title_short |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| title_full |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| title_fullStr |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| title_full_unstemmed |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| title_sort |
Fine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniques |
| dc.creator.fl_str_mv |
Avalos López, Fernando Andrés |
| dc.contributor.advisor.none.fl_str_mv |
Manrique Piramanrique, Rubén Francisco |
| dc.contributor.author.none.fl_str_mv |
Avalos López, Fernando Andrés |
| dc.subject.keyword.eng.fl_str_mv |
Face anti-spoofing |
| topic |
Face anti-spoofing Biometric privacy-enhancing techniques Fine-tuning Performance analysis Ingeniería |
| dc.subject.keyword.none.fl_str_mv |
Biometric privacy-enhancing techniques Fine-tuning Performance analysis |
| dc.subject.themes.spa.fl_str_mv |
Ingeniería |
| description |
Face Anti-Spoofing (FAS) systems are entrusted with the task of determining whether the content of a facial image is genuine or fake. The performance and overall quality of those systems depend on the data they are fed during their development stage, meaning that high-resolution and diverse facial datasets are commonly used. Even though FAS systems are particularly useful in some situations, they do raise concerns in regards to the privacy of the individuals whose images are used to train them. The straightforward solution is to apply filters to the facial images, at the expense of the systems’ performance, since recognising faces becomes increasingly harder. This is where Biometric Privacy-Enhancing Techniques (B-PETs) come into play, which help to alleviate the adversarial tension between biometric utility and privacy gains. This work concerned itself with assessing the impact of 3 distinct image-level B-PETs in the performance of 3 architecturally distinct FAS systems and found that image-level B-PETs are not fit for finding a valuable trade-off, suggesting that more sophisticated techniques are needed. |
| publishDate |
2024 |
| dc.date.accessioned.none.fl_str_mv |
2024-02-02T20:35:28Z |
| dc.date.issued.none.fl_str_mv |
2024-02-01 |
| dc.date.accepted.none.fl_str_mv |
2024-02-01 |
| dc.date.available.none.fl_str_mv |
2025-01-31 |
| dc.type.none.fl_str_mv |
Trabajo de grado - Pregrado |
| dc.type.driver.none.fl_str_mv |
info:eu-repo/semantics/bachelorThesis |
| dc.type.version.none.fl_str_mv |
info:eu-repo/semantics/acceptedVersion |
| dc.type.coar.none.fl_str_mv |
http://purl.org/coar/resource_type/c_7a1f |
| dc.type.content.none.fl_str_mv |
Text |
| dc.type.redcol.none.fl_str_mv |
http://purl.org/redcol/resource_type/TP |
| format |
http://purl.org/coar/resource_type/c_7a1f |
| status_str |
acceptedVersion |
| dc.identifier.uri.none.fl_str_mv |
https://hdl.handle.net/1992/73851 |
| dc.identifier.instname.none.fl_str_mv |
instname:Universidad de los Andes |
| dc.identifier.reponame.none.fl_str_mv |
reponame:Repositorio Institucional Séneca |
| dc.identifier.repourl.none.fl_str_mv |
repourl:https://repositorio.uniandes.edu.co/ |
| url |
https://hdl.handle.net/1992/73851 |
| identifier_str_mv |
instname:Universidad de los Andes reponame:Repositorio Institucional Séneca repourl:https://repositorio.uniandes.edu.co/ |
| dc.language.iso.none.fl_str_mv |
eng |
| language |
eng |
| dc.relation.references.none.fl_str_mv |
[1] Face anti-spoofing, face presentation attack detection. https://cvlab.cse.msu.edu/project-face-anti.html, 2022. [2] Face liveness detection anti-spoofing web app. https://github.com/birdowl21/Face-Liveness-Detection-Anti-Spoofing-Web-App, 2022. [3] Minivision AI. Silent face anti-spoofing. https://github.com/minivision-ai/Silent-Face-Anti-Spoofing, 2020. [4] Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge. A neural algorithm of artistic style, 2015. [5] A.S. Georghiades, P.N. Belhumeur, and D.J. Kriegman. From few to many: illumination cone models for face recognition under variable lighting and pose. IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(6):643–660, 2001. DOI 10.1109/34.927464. [6] Gary B. Huang, Marwan Mattar, Honglak Lee, and Erik Learned-Miller. Learning to align from scratch. In NIPS, 2012. [7] Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks, 2019. [8] Stan Z. Li. Encyclopedia of Biometrics. Springer Publishing Company, Incorporated, 1st edition, 2009. ISBN 0387730028. [9] Blaž Meden, Peter Rot, Philipp Terhörst, Naser Damer, Arjan Kuijper, Walter J. Scheirer, Arun Ross, Peter Peer, and Vitomir Štruc. Privacyenhancing face biometrics: A comprehensive survey. Trans. Info. For. Sec., 16:4147–4183, jan 2021. ISSN 1556-6013. DOI 10.1109/TIFS.2021.3096024. URL https://doi.org/10.1109/TIFS.2021.3096024. [10] Beate Roessler and Judith DeCew. Privacy. In Edward N. Zalta and Uri Nodelman, editors, The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University, Winter 2023 edition, 2023. [11] C. Vega Fernández. Estrategias para la generación sintética de imágenes y su aplicación a escenarios de aumentación de datos en el desarrollo de sistemas face anti-spoofing, 2023. [12] Di Wen, Hu Han, and Anil K. Jain. Face spoof detection with image distortion analysis. IEEE Transactions on Information Forensics and Security, 10(4):746–761, 2015. DOI 10. 1109/TIFS.2015.2400395. |
| dc.rights.en.fl_str_mv |
Attribution 4.0 International |
| dc.rights.uri.none.fl_str_mv |
http://creativecommons.org/licenses/by/4.0/ |
| dc.rights.accessrights.none.fl_str_mv |
info:eu-repo/semantics/embargoedAccess |
| dc.rights.coar.none.fl_str_mv |
http://purl.org/coar/access_right/c_f1cf |
| rights_invalid_str_mv |
Attribution 4.0 International http://creativecommons.org/licenses/by/4.0/ http://purl.org/coar/access_right/c_f1cf |
| eu_rights_str_mv |
embargoedAccess |
| dc.format.extent.none.fl_str_mv |
48 páginas |
| dc.format.mimetype.none.fl_str_mv |
application/pdf |
| dc.publisher.none.fl_str_mv |
Universidad de los Andes |
| dc.publisher.program.none.fl_str_mv |
Ingeniería de Sistemas y Computación |
| dc.publisher.faculty.none.fl_str_mv |
Facultad de Ingeniería |
| dc.publisher.department.none.fl_str_mv |
Departamento de Ingeniería Sistemas y Computación |
| publisher.none.fl_str_mv |
Universidad de los Andes |
| institution |
Universidad de los Andes |
| bitstream.url.fl_str_mv |
https://repositorio.uniandes.edu.co/bitstreams/09246f05-51a7-4831-b1c1-ae3fba9403bc/download https://repositorio.uniandes.edu.co/bitstreams/065a7a78-845c-4c7d-8afe-0b94cca66e0c/download https://repositorio.uniandes.edu.co/bitstreams/76601e02-325b-4111-b282-3f813b3c04cc/download https://repositorio.uniandes.edu.co/bitstreams/07ec10ca-8552-42d8-aa3d-3f76b6bc7b09/download https://repositorio.uniandes.edu.co/bitstreams/a3254165-720f-4d92-8226-e3a7b54d7ac2/download https://repositorio.uniandes.edu.co/bitstreams/6aaede7c-9de7-4033-b15c-056a8fa4af87/download https://repositorio.uniandes.edu.co/bitstreams/6d24a3d6-25b8-4ec7-a7e2-308b87a4fdef/download https://repositorio.uniandes.edu.co/bitstreams/6e9f124a-a94c-43c9-bbf2-0b8eaec6dfdc/download |
| bitstream.checksum.fl_str_mv |
ae9e573a68e7f92501b6913cc846c39f 20e377555832c95bcd8be056bfa6bcd8 31acccf1bc77550f91010ca52a45ad73 0175ea4a2d4caec4bbcc37e300941108 c590b6fe9e631781067acd7971a0508c 530d320eb9a3b695b3582e4fc981c68e a4b2b6ae24fc9f640384a3ae25136f4a 945a639b4228c3721ab369ad4ff8237c |
| bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 MD5 MD5 MD5 |
| repository.name.fl_str_mv |
Repositorio institucional Séneca |
| repository.mail.fl_str_mv |
adminrepositorio@uniandes.edu.co |
| _version_ |
1808390177730592768 |
| spelling |
Manrique Piramanrique, Rubén Franciscovirtual::312-1Avalos López, Fernando Andrés2024-02-02T20:35:28Z2025-01-312024-02-012024-02-01https://hdl.handle.net/1992/73851instname:Universidad de los Andesreponame:Repositorio Institucional Sénecarepourl:https://repositorio.uniandes.edu.co/Face Anti-Spoofing (FAS) systems are entrusted with the task of determining whether the content of a facial image is genuine or fake. The performance and overall quality of those systems depend on the data they are fed during their development stage, meaning that high-resolution and diverse facial datasets are commonly used. Even though FAS systems are particularly useful in some situations, they do raise concerns in regards to the privacy of the individuals whose images are used to train them. The straightforward solution is to apply filters to the facial images, at the expense of the systems’ performance, since recognising faces becomes increasingly harder. This is where Biometric Privacy-Enhancing Techniques (B-PETs) come into play, which help to alleviate the adversarial tension between biometric utility and privacy gains. This work concerned itself with assessing the impact of 3 distinct image-level B-PETs in the performance of 3 architecturally distinct FAS systems and found that image-level B-PETs are not fit for finding a valuable trade-off, suggesting that more sophisticated techniques are needed.Ingeniero de Sistemas y ComputaciónPregrado48 páginasapplication/pdfengUniversidad de los AndesIngeniería de Sistemas y ComputaciónFacultad de IngenieríaDepartamento de Ingeniería Sistemas y ComputaciónAttribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/embargoedAccesshttp://purl.org/coar/access_right/c_f1cfFine-tuning face anti-spoofing models: exploring the transformative impact of image-level biometric privacy-enhancing techniquesTrabajo de grado - Pregradoinfo:eu-repo/semantics/bachelorThesisinfo:eu-repo/semantics/acceptedVersionhttp://purl.org/coar/resource_type/c_7a1fTexthttp://purl.org/redcol/resource_type/TPFace anti-spoofingBiometric privacy-enhancing techniquesFine-tuningPerformance analysisIngeniería[1] Face anti-spoofing, face presentation attack detection. https://cvlab.cse.msu.edu/project-face-anti.html, 2022.[2] Face liveness detection anti-spoofing web app. https://github.com/birdowl21/Face-Liveness-Detection-Anti-Spoofing-Web-App, 2022.[3] Minivision AI. Silent face anti-spoofing. https://github.com/minivision-ai/Silent-Face-Anti-Spoofing, 2020.[4] Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge. A neural algorithm of artistic style, 2015.[5] A.S. Georghiades, P.N. Belhumeur, and D.J. Kriegman. From few to many: illumination cone models for face recognition under variable lighting and pose. IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(6):643–660, 2001. DOI 10.1109/34.927464.[6] Gary B. Huang, Marwan Mattar, Honglak Lee, and Erik Learned-Miller. Learning to align from scratch. In NIPS, 2012.[7] Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks, 2019.[8] Stan Z. Li. Encyclopedia of Biometrics. Springer Publishing Company, Incorporated, 1st edition, 2009. ISBN 0387730028.[9] Blaž Meden, Peter Rot, Philipp Terhörst, Naser Damer, Arjan Kuijper, Walter J. Scheirer, Arun Ross, Peter Peer, and Vitomir Štruc. Privacyenhancing face biometrics: A comprehensive survey. Trans. Info. For. Sec., 16:4147–4183, jan 2021. ISSN 1556-6013. DOI 10.1109/TIFS.2021.3096024. URL https://doi.org/10.1109/TIFS.2021.3096024.[10] Beate Roessler and Judith DeCew. Privacy. In Edward N. Zalta and Uri Nodelman, editors, The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University, Winter 2023 edition, 2023.[11] C. Vega Fernández. Estrategias para la generación sintética de imágenes y su aplicación a escenarios de aumentación de datos en el desarrollo de sistemas face anti-spoofing, 2023.[12] Di Wen, Hu Han, and Anil K. Jain. Face spoof detection with image distortion analysis. IEEE Transactions on Information Forensics and Security, 10(4):746–761, 2015. DOI 10. 1109/TIFS.2015.2400395.201911468Publication9f6e12e0-098e-4548-ab81-75552e8385e7virtual::312-19f6e12e0-098e-4548-ab81-75552e8385e7virtual::312-1LICENSElicense.txtlicense.txttext/plain; charset=utf-82535https://repositorio.uniandes.edu.co/bitstreams/09246f05-51a7-4831-b1c1-ae3fba9403bc/downloadae9e573a68e7f92501b6913cc846c39fMD51ORIGINALFine-Tuning Face Anti-Spoofing Models.pdfFine-Tuning Face Anti-Spoofing Models.pdfDocument access is restricted for one year in case it ends up being published in a journal or conference.application/pdf8531657https://repositorio.uniandes.edu.co/bitstreams/065a7a78-845c-4c7d-8afe-0b94cca66e0c/download20e377555832c95bcd8be056bfa6bcd8MD52autorizacion tesis fernando signed.pdfautorizacion tesis fernando signed.pdfHIDEapplication/pdf241144https://repositorio.uniandes.edu.co/bitstreams/76601e02-325b-4111-b282-3f813b3c04cc/download31acccf1bc77550f91010ca52a45ad73MD53CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8908https://repositorio.uniandes.edu.co/bitstreams/07ec10ca-8552-42d8-aa3d-3f76b6bc7b09/download0175ea4a2d4caec4bbcc37e300941108MD54TEXTFine-Tuning Face Anti-Spoofing Models.pdf.txtFine-Tuning Face Anti-Spoofing Models.pdf.txtExtracted texttext/plain31910https://repositorio.uniandes.edu.co/bitstreams/a3254165-720f-4d92-8226-e3a7b54d7ac2/downloadc590b6fe9e631781067acd7971a0508cMD55autorizacion tesis fernando signed.pdf.txtautorizacion tesis fernando signed.pdf.txtExtracted texttext/plain1176https://repositorio.uniandes.edu.co/bitstreams/6aaede7c-9de7-4033-b15c-056a8fa4af87/download530d320eb9a3b695b3582e4fc981c68eMD57THUMBNAILFine-Tuning Face Anti-Spoofing Models.pdf.jpgFine-Tuning Face Anti-Spoofing Models.pdf.jpgGenerated Thumbnailimage/jpeg3561https://repositorio.uniandes.edu.co/bitstreams/6d24a3d6-25b8-4ec7-a7e2-308b87a4fdef/downloada4b2b6ae24fc9f640384a3ae25136f4aMD56autorizacion tesis fernando signed.pdf.jpgautorizacion tesis fernando signed.pdf.jpgGenerated Thumbnailimage/jpeg11165https://repositorio.uniandes.edu.co/bitstreams/6e9f124a-a94c-43c9-bbf2-0b8eaec6dfdc/download945a639b4228c3721ab369ad4ff8237cMD581992/73851oai:repositorio.uniandes.edu.co:1992/738512024-02-16 14:41:16.587http://creativecommons.org/licenses/by/4.0/Attribution 4.0 Internationalembargohttps://repositorio.uniandes.edu.coRepositorio institucional Sénecaadminrepositorio@uniandes.edu.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 |