Abstract:
With the enormous amount of data, especially images, getting produced every minute and the ease of availability of cloud services at cheaper rates, there is an exponential increase in the amount of data being moved to the third-party cloud data centers. However, these data centers and the transmission of data are vulnerable to various intruding attacks resulting in untoward breaching of security and privacy. To avoid this, the visual information of image data is encrypted before transmission and storage. Albeit, performing image processing operations on these encrypted images is a challenging task, but at the same time, it saves computational and transmission costs as compared to computing those operations in other domains. This report investigates and explores privacy preserving face search over cloud. Particularly, given a face image as query, we attempt to find its region among encrypted images. We propose a novel privacy preserving face search model that extracts features from face image and identifies the face object region confined in the encrypted image
