Human-Imperceptible Identification with Learnable Lensless Imaging
This work addresses privacy protection in imaging systems for applications like surveillance or personal devices, but it is incremental as it builds on existing lensless imaging methods.
The paper tackles the trade-off between visual privacy and recognition accuracy in lensless imaging by proposing a learnable framework that uses loss functions based on total variation, invertibility, and the restricted isometry property to make images imperceptible to humans while maintaining identification accuracy, validated through subjective evaluation and hardware implementation.
Lensless imaging protects visual privacy by capturing heavily blurred images that are imperceptible for humans to recognize the subject but contain enough information for machines to infer information. Unfortunately, protecting visual privacy comes with a reduction in recognition accuracy and vice versa. We propose a learnable lensless imaging framework that protects visual privacy while maintaining recognition accuracy. To make captured images imperceptible to humans, we designed several loss functions based on total variation, invertibility, and the restricted isometry property. We studied the effect of privacy protection with blurriness on the identification of personal identity via a quantitative method based on a subjective evaluation. Moreover, we validate our simulation by implementing a hardware realization of lensless imaging with photo-lithographically printed masks.