Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking

Anton Mølbjerg Eskildsen; Dan Witzner Hansen

doi:10.1145/3448017.3457385

Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking

Anton Mølbjerg Eskildsen, Dan Witzner Hansen

Machine Learning

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

We present a framework to model and evaluate obfuscation methods for removing sensitive information in eye-tracking. The focus is on preventing iris-pattern identification. Candidate methods have to be effective at removing information while retaining high utility for gaze estimation. We propose several obfuscation methods that drastically outperform existing ones. A stochastic grid-search is used to determine optimal method parameters and evaluate the model framework.
Precise obfuscation and gaze effects are measured for selected parameters.
Two attack scenarios are considered and evaluated. We show that large datasets are susceptible to probabilistic attacks, even with seemingly effective obfuscation methods. However, additional data is needed to more accurately access the probabilistic security.

Original language	English
Title of host publication	ETRA '21 Full Papers : ACM Symposium on Eye Tracking Research and Applications
Number of pages	10
Publisher	Association for Computing Machinery
Publication date	25 May 2021
Pages	1-10
Article number	2
ISBN (Electronic)	978-1-4503-8344-8
DOIs	https://doi.org/10.1145/3448017.3457385
Publication status	Published - 25 May 2021
Event	ACM Symposium on Eye Tracking Research & Applications - Online Duration: 24 May 2021 → 27 May 2021 https://etra.acm.org/2021/

Conference

Conference	ACM Symposium on Eye Tracking Research & Applications
Location	Online
Period	24/05/2021 → 27/05/2021
Internet address	https://etra.acm.org/2021/

Keywords

obfuscation
eye-tracking
privacy

Access to Document

10.1145/3448017.3457385

iris_obfuscationAccepted author manuscript, 1.01 MB

Cite this

@inproceedings{baa5d9950f21430bb58986491fdd6599,

title = "Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking",

abstract = "We present a framework to model and evaluate obfuscation methods for removing sensitive information in eye-tracking. The focus is on preventing iris-pattern identification. Candidate methods have to be effective at removing information while retaining high utility for gaze estimation. We propose several obfuscation methods that drastically outperform existing ones. A stochastic grid-search is used to determine optimal method parameters and evaluate the model framework. Precise obfuscation and gaze effects are measured for selected parameters. Two attack scenarios are considered and evaluated. We show that large datasets are susceptible to probabilistic attacks, even with seemingly effective obfuscation methods. However, additional data is needed to more accurately access the probabilistic security.",

keywords = "obfuscation, eye-tracking, privacy, obfuscation, eye-tracking, privacy",

author = "Eskildsen, {Anton M{\o}lbjerg} and Hansen, {Dan Witzner}",

year = "2021",

month = may,

day = "25",

doi = "10.1145/3448017.3457385",

language = "English",

pages = "1--10",

booktitle = "ETRA '21 Full Papers",

publisher = "Association for Computing Machinery",

address = "United States",

note = "ACM Symposium on Eye Tracking Research & Applications ; Conference date: 24-05-2021 Through 27-05-2021",

url = "https://etra.acm.org/2021/",

}

Eskildsen, AM & Hansen, DW 2021, Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking. in ETRA '21 Full Papers: ACM Symposium on Eye Tracking Research and Applications., 2, Association for Computing Machinery, pp. 1-10, ACM Symposium on Eye Tracking Research & Applications, 24/05/2021. https://doi.org/10.1145/3448017.3457385

Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking. / Eskildsen, Anton Mølbjerg; Hansen, Dan Witzner.
ETRA '21 Full Papers: ACM Symposium on Eye Tracking Research and Applications. Association for Computing Machinery, 2021. p. 1-10 2.

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Analysis of iris obfuscation

T2 - ACM Symposium on Eye Tracking Research & Applications

AU - Eskildsen, Anton Mølbjerg

AU - Hansen, Dan Witzner

PY - 2021/5/25

Y1 - 2021/5/25

N2 - We present a framework to model and evaluate obfuscation methods for removing sensitive information in eye-tracking. The focus is on preventing iris-pattern identification. Candidate methods have to be effective at removing information while retaining high utility for gaze estimation. We propose several obfuscation methods that drastically outperform existing ones. A stochastic grid-search is used to determine optimal method parameters and evaluate the model framework. Precise obfuscation and gaze effects are measured for selected parameters. Two attack scenarios are considered and evaluated. We show that large datasets are susceptible to probabilistic attacks, even with seemingly effective obfuscation methods. However, additional data is needed to more accurately access the probabilistic security.

AB - We present a framework to model and evaluate obfuscation methods for removing sensitive information in eye-tracking. The focus is on preventing iris-pattern identification. Candidate methods have to be effective at removing information while retaining high utility for gaze estimation. We propose several obfuscation methods that drastically outperform existing ones. A stochastic grid-search is used to determine optimal method parameters and evaluate the model framework. Precise obfuscation and gaze effects are measured for selected parameters. Two attack scenarios are considered and evaluated. We show that large datasets are susceptible to probabilistic attacks, even with seemingly effective obfuscation methods. However, additional data is needed to more accurately access the probabilistic security.

KW - obfuscation

KW - eye-tracking

KW - privacy

KW - obfuscation

KW - eye-tracking

KW - privacy

U2 - 10.1145/3448017.3457385

DO - 10.1145/3448017.3457385

M3 - Article in proceedings

SP - 1

EP - 10

BT - ETRA '21 Full Papers

PB - Association for Computing Machinery

Y2 - 24 May 2021 through 27 May 2021

ER -

Analysis of iris obfuscation: Generalising eye information processes for privacy studies in eye tracking

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this