Your Tattletale Gait Privacy Invasiveness of IMU Gait Data

Rasnayaka, Sanka; Sim, Terence

doi:10.1109/ijcb48548.2020.9304922

Cited by 8 publications

(4 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We then analysed the privacy invasiveness in our implementation for images with mask and without mask, to understand the privacy preservation when using a face mask. We show that there is no significant difference in privacy protection by quantifying the privacy invasiveness using the Privacy Vulnerability Index (PVI) [25] for both settings, which recorded only a 2.9% difference that implies no significance in wearing a mask.…”

Section: Introductionmentioning

confidence: 89%

“…The use of biometrics have raised various privacy concerns due to the possibility of predicting protected attributes. Many studies have evaluated the predictability of soft-biometric attributes such as age, gender and race using common biometrics such as face [14], iris [28], fingerprint [3], voice [10] and gait [25]. In this work, we go beyond than prediction and provide means of quantifying the privacy invasiveness in systems that use soft-biometric.…”

Section: Biometrics and Privacymentioning

confidence: 99%

“…The Privacy Vulnerability Index [25] is used to quantify the privacy invasiveness of a biometric modality. We use this measure to compare the privacy invasiveness of face images and masked face images.…”

Section: Privacy Vulnerability Index(pvi)mentioning

confidence: 99%

See 2 more Smart Citations

Does a Face Mask Protect my Privacy?: Deep Learning to Predict Protected Attributes from Masked Face Images

Seneviratne,

Kasthuriarachchi,

Rasnayaka

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Contactless and efficient systems are implemented rapidly to advocate preventive methods in the fight against the COVID-19 pandemic. Despite the positive benefits of such systems, there is potential for exploitation by invading user privacy. In this work, we analyse the privacy invasiveness of face biometric systems by predicting privacysensitive soft-biometrics using masked face images. We train and apply a CNN based on the ResNet-50 architecture with 20,003 synthetic masked images and measure the privacy invasiveness. Despite the popular belief of the privacy benefits of wearing a mask among people, we show that there is no significant difference to privacy invasiveness when a mask is worn. In our experiments we were able to accurately predict sex (94.7%), race (83.1%) and age (MAE 6.21 and RMSE 8.33) from masked face images. Our proposed approach can serve as a baseline utility to evaluate the privacy-invasiveness of artificial intelligence systems that make use of privacy-sensitive information. We open-source all contributions for reproducibility and broader use by the research community.

show abstract

Section: Introductionmentioning

confidence: 89%

Section: Biometrics and Privacymentioning

confidence: 99%

See 1 more Smart Citation

Does a Face Mask Protect my Privacy?: Deep Learning to Predict Protected Attributes from Masked Face Images

Seneviratne,

Kasthuriarachchi,

Rasnayaka

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For the text reading task (Task 2) and for the tapping task (Task 4), only accelerometer and gyroscope data are considered employing rolling time windows of 32 samples to be fed to a separate Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) [19]. For Task 2, a N-class classifier is implemented for N identities (N = 51, the users in the training set).…”

Section: The Nus-uoa-uom Teammentioning

confidence: 99%

IJCB 2022 Mobile Behavioral Biometrics Competition (MobileB2C)

Stragapede¹,

Vera-Rodríguez²,

Tolosana³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper describes the experimental framework and results of the IJCB 2022 Mobile Behavioral Biometrics Competition (MobileB2C). The aim of MobileB2C is benchmarking mobile user authentication systems based on behavioral biometric traits transparently acquired by mobile devices during ordinary Human-Computer Interaction (HCI), using a novel public database, Be-havePassDB 1 , and a standard experimental protocol. The competition is divided into four tasks corresponding to typical user activities: keystroke, text reading, gallery swiping, and tapping. The data are composed of touchscreen data and several background sensor data simultaneously acquired. "Random" (different users with different devices) and "skilled" (different user on the same device attempting to imitate the legitimate one) impostor scenarios are considered. The results achieved by the participants show the feasibility of user authentication through behavioral biometrics, although this proves to be a nontrivial challenge. MobileB2C will be established as an on-going competition 2 .

show abstract