Privacy Preserving Gaze Estimation using Synthetic Images via a Randomized Encoding Based Framework

被引:14
作者
Bozkir, Efe [1 ]
Uenal, Ali Burak [2 ]
Akguen, Mete [3 ]
Kasneci, Enkelejda [1 ]
Pfeifer, Nico [2 ]
机构
[1] Univ Tubingen, Human Comp Interact, Tubingen, Germany
[2] Univ Tubingen, Methods Med Informat, Tubingen, Germany
[3] Univ Tubingen, Methods Med Informat, Translat Bioinformat, Tubingen, Germany
来源
ETRA 2020 SHORT PAPERS: ACM SYMPOSIUM ON EYE TRACKING RESEARCH & APPLICATIONS | 2020年
关键词
privacy preserving machine learning; gaze estimation; randomized encoding; eye tracking; human computer interaction;
D O I
10.1145/3379156.3391364
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Eye tracking is handled as one of the key technologies for applications that assess and evaluate human attention, behavior, and biometrics, especially using gaze, pupillary, and blink behaviors. One of the challenges with regard to the social acceptance of eye tracking technology is however the preserving of sensitive and personal information. To tackle this challenge, we employ a privacy-preserving framework based on randomized encoding to train a Support Vector Regression model using synthetic eye images privately to estimate the human gaze. During the computation, none of the parties learn about the data or the result that any other party has. Furthermore, the party that trains the model cannot reconstruct pupil, blinks or visual scanpath. The experimental results show that our privacy-preserving framework is capable of working in real-time, with the same accuracy as compared to non-private version and could be extended to other eye tracking related problems.
引用
收藏
页数:5
相关论文
共 27 条
[1]   Just Gaze and Wave: Exploring the Use of Gaze and Gestures for Shoulder-surfing Resilient Authentication [J].
Abdrabou, Yasmeen ;
Khamis, Mohamed ;
Eisa, Rana Mohamed ;
Ismail, Sherif ;
Elmougy, Amrl .
ETRA 2019: 2019 ACM SYMPOSIUM ON EYE TRACKING RESEARCH & APPLICATIONS, 2019,
[2]  
[Anonymous], 2010, P S EYE TRACK RES AP, DOI DOI 10.1145/1743666.1743679
[3]   Cross-subject workload classification using pupil-related measures [J].
Appel, Tobias ;
Scharinger, Christian ;
Gerjets, Peter ;
Kasneci, Enkelejda .
2018 ACM SYMPOSIUM ON EYE TRACKING RESEARCH & APPLICATIONS (ETRA 2018), 2018,
[4]   Cryptography in NC0 [J].
Applebaum, Benny ;
Ishai, Yuval ;
Kushilevitz, Eyal .
SIAM JOURNAL ON COMPUTING, 2006, 36 (04) :845-888
[5]   Computationally private randomizing polynomials and their applications [J].
Applebaum, Benny ;
Ishai, Yuval ;
Kushilevitz, Eyal .
COMPUTATIONAL COMPLEXITY, 2006, 15 (02) :115-162
[6]  
Applebaum B, 2017, INFORM SEC CRYPT TEX, P1, DOI 10.1007/978-3-319-57048-8_1
[7]   Detecting Personality Traits Using Eye-Tracking Data [J].
Berkovsky, Shlomo ;
Taib, Ronnie ;
Koprinska, Irena ;
Wang, Eileen ;
Zeng, Yucheng ;
Li, Jingjie ;
Kleitman, Sabina .
CHI 2019: PROCEEDINGS OF THE 2019 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS, 2019,
[8]   Defending Yarbus: Eye movements reveal observers' task [J].
Borji, Ali ;
Itti, Laurent .
JOURNAL OF VISION, 2014, 14 (03)
[9]   Assessment of Driver Attention during a Safety Critical Situation in VR to Generate VR-based Training [J].
Bozkir, Efe ;
Geisler, David ;
Kasneci, Enkelejda .
ACM CONFERENCE ON APPLIED PERCEPTION (SAP 2019), 2019,
[10]   Online Recognition of Driver-Activity Based on Visual Scanpath Classification [J].
Braunagel, Christian ;
Geisler, David ;
Rosenstiel, Wolfgang ;
Kasneci, Enkelejda .
IEEE INTELLIGENT TRANSPORTATION SYSTEMS MAGAZINE, 2017, 9 (04) :23-36