Semester 1: Build the algorithm of a 2D model from a video surveillance camera for motion detection. 
Semester 2: Apply heterogeneous parallel processing to compute and to display.  

Motion detection can refer to humans, vehicles, animals, falling leaves, but excludes change of environmental lighting intensity.  Our model shall be capable of knowing the type of moving objects and types of simultaneous motions.

The university is a world leader in Computer Vision and Multimedia Imaging.  Compucon co-supervises a 4th year student with the university to learn about computer based data modelling techniques among other tools to achieve effects that were technically not achievable before.  Commercial off-the-shelf video surveillance cameras are very capable of recording images in computer hard disks at high resolutions (such as 1920 x 1024) and frame rates (such as 30 frames per second).  They are also capable of applying motion detection criteria to reduce the bandwidth for data transmission and recording.  However these cameras did not understand the meaning of patterns or the environment as seen and recorded.  This is where computer data modelling comes in.  Data modelling will attempt to give meanings to images and environment to the recording.  This in turn will allow more meaningful actions to be taken and thus a higher level of automation which we called artificial intelligence. 

The university is also proactive in the research of high performance computing.  We expect the above described data model to require a lot of computation power to work out the solutions.  More to the point, the computation has to be done in real time so that automation happens in real time.  This means computation in as little time as possible and amounts to high performance computing.  The university has a high performance computing research group and possess a large amount of expertise in this area.

This project will involve both areas of university expertise to help the student undertaking this project.  The practical experience to be gained from this project will highly consolidate knowledge of the 2 areas learned by the student over the course of the degree.

At the same time, the industry supervisor will contribute significantly to this project in 2 respects.  The industry supervisor is Compucon New Zealand.  It is a computer making operation in New Zealand with a high level of engineering expertise.  The student can expect to learn from a Chartered Engineer and several senior graduates of the university.   In addition to computer making, Compucon is also a total solution developer for Internet Protocol based video surveillance and has implemented a few medium scale projects in New Zealand for commercial customers.

Industry Supervisor: TN Chan
Academic Supervisor: Professor Reinhard Klette
Program Coordinator: Dr Mano