Privacy-Preserving Image Acquisition Using Trainable Optical Kernel

Abstract

Preserving privacy is a growing concern in our society where sensors and in particular cameras are ubiquitous. In this work, for the first time, we propose a trainable image acquisition method that removes the sensitive identity revealing information in the optical domain before it reaches the image sensor. The method benefits from a trainable optical convolution kernel which transmits the desired information while filters out the sensitive information. As the sensitive information is suppressed before it reaches the image sensor, it does not enter the digital domain therefore is unretrievable by any sort of privacy attack. This is in contrast with the current digital privacy-preserving methods that are all vulnerable to direct access attack. Also, in contrast with the previous optical privacy-preserving methods that cannot be trained, our method is data-driven and optimized for the specific application at hand. Moreover, there is no additional computation, memory, or power burden on the acquisition system since this processing happens passively in the optical domain and can even be used together and on top of the fully digital privacy-preserving systems. The proposed approach is generic and adaptable to different digital neural networks, and desired and sensitive content pairs. We demonstrate our new method for several scenarios such as smile detection as the desired attribute while the gender is filtered out as the sensitive content. We trained the optical kernel in conjunction with two adversarial neural networks where the analysis network tries to detect the desired attribute and the adversarial network tries to detect the sensitive content. We show that this method is able to reduce 65.1% of sensitive content when it is selected to be the gender and as the kernel is optimized, it only causes 7.3% reduction of the desired information content. Moreover, we reconstruct the original faces using the deep image reconstruction method that confirms the ineffectiveness of reconstruction attacks to obtain the sensitive content.