Date:

Privacy protected video surveillance

Abstract: The proliferation of digital cameras, portable storage devices and wireless networks make sharing of multimedia data easier than ever. Such casual exchange of data, however, has increasingly raised questions on how sensitive multimedia information can be protected. How can employees' right to privacy be protected when numerous surveillance cameras are installed in their work place? How can a consumer protect his/her private pictures sent to a commercial photo laboratory for processing? How can organizations share multimedia intelligence without divulging sensitive information? While significant progress has been made in extracting useful information and pattern from multimedia contents, little work has been done to investigate how sensitive multimedia information can be protected.

In this talk, I will describe some of our recent works in privacy protection of multimedia information. The main focus will be on
privacy protection in a multi-camera video surveillance system. There are three technical challenges in developing such a system: How to identify selected individuals? How to obfuscate their appearances while keeping the video useful for other purposes? How to authenticate such obfuscation or even to reverse the obfuscation given proper authorization? Our contributions in solving these problems are based on an optimal design of a multi-camera network for people identification, an efficient object-based video inpainting algorithm for complete object erasure and a secure distributed privacy data management system based on software agents and rate-distortion optimized data hiding. If time permits, I will also discuss our latest works in audio privacy protection and image processing in encrypted domain.

Date:

Phonon-Energy-Coupling Enhancement- A New Physics Effect: Leakage Current Reduction of Gate Dielectrics

Abstract: Pending.

Date:

Title to be announced

Abstract: Pending.

Date:

Title to be announced

Abstract: Pending.