PROJECT TITLE:

Privacy-Preserving Outsourced Media Search - 2016

ABSTRACT:

This work proposes a privacy-protection framework for an necessary application referred to as outsourced media search. This scenario involves a information owner, a client, and an untrusted server, where the owner outsources an enquiry service to the server. Thanks to lack of trust, the privacy of the client and the owner ought to be protected. The framework depends on multimedia hashing and symmetric encryption. It needs concerned parties to participate in a very privacy-enhancing protocol. Extra processing steps are dispensed by the owner and the client: (i) before outsourcing low-level media features to the server, the owner has to one-manner hash them, and partially encrypt every hash-price; (ii) the client completes the similarity search by re-ranking the most similar candidates received from the server. One-approach hashing and encryption add ambiguity to data and build it troublesome for the server to infer contents from database items and queries, so the privacy of each the owner and also the client is enforced. The proposed framework realizes trade-offs among strength of privacy enforcement, quality of search, and complexity, as a result of the knowledge loss can be tuned throughout hashing and encryption. Intensive experiments demonstrate the effectiveness and the flexibleness of the framework.