How to read faces without looking at them
This addresses privacy concerns for users in emotion-sensitive scenarios, offering an incremental improvement over existing methods.
The paper tackles the privacy issue in facial emotion recognition by proposing a compressive analysis system that acquires faces in a way that prevents usable reconstruction while maintaining adjustable inference accuracy, achieving a 15% reduction in privacy risk with 85% recognition accuracy.
Face reading is the most intuitive aspect of emotion recognition. Unfortunately, digital analysis of facial expression requires digitally recording personal faces. As emotional analysis is particularly required in a more poised scenario, capturing faces becomes a gross violation of privacy. In this paper, we use the concept of compressive analysis to conceptualise a system which compressively acquires faces in order to ascertain unusable reconstruction, while allowing for acceptable (and adjustable) accuracy in inference.