[mobile system] I-Pic: A Platform for Privacy-Compliant Image Capture

PDF

Summary

I-Pic is a trusted software platform that integrates digital capture with user-defined privacy. User choose a level of privacy based upon social context. Privacy policies are advertised via Bluetooth.

Architecture

Workflow

Design Goals

Authors conducted an online survey to support their design goals

1. Privacy policies should be individualized
2. Privacy policies should be situational
3. Compliance by courtesy is sufficient

Techniques

1. Face detection: HeadHunter (high recall, low precision)
2. Face recognition: DNN based, Imagenet finetuned feature extraction (128 dimension) + SVM
3. Secure Matching Protocol: secure dot product + gabled circuits

Evaluation

15 volunteers carried ipic compliant devices with specified policies of when and where to show or blur their faces

Faces Collected

Result

Methodology of Analysis

1. Analyzed visual pipeline failure by categorizing 1000+ collected images into different categories (good lighting, occluded…) and gives recall rate for each category.
2. Mapped categories above into social events
3. Compared authors’ face detection with others on recall rate.
4. Analyzed computation, bandwidth, and energy cost of secure matching protocol