Model-Based Detection of Emergent Behavior in Distributed Systems

The lack of central control in distributed systems makes the analysis and design of such systems a challenging task.  In fact many faults are introduced into the system at this stage of the software development life cycle.  Therefore discovering and removing design validation of distributed software systems prior to the implementation phase is greatly desirable as it results in huge savings in time and cost.  An effective and efficient approach to describing the requirements of distributed systems is by using scenarios.  Scenarios are commonly represented using message sequence charts or se

Accelerating Automated Synchronization of Multiple-Camera Videos on multicore platforms.

This project aims to explore the feasibility of and the tradeoffs involved with accelerating multi-camera synchronization algorithms on multicore processors. Specifically, this project plans to explore acceleration using ‘traditional’ multicore CPUs and massively-parallel Graphics Processing Units (GPUs). PluralEyes, a product for synchronizing audio and video recordings released by Singular Software, a Vancouver-area company, is the software application informing this research and stands to directly benefit from the insights gained. 

Phase II – A Software Platform for Generation of Mutually Correlated Multi-Dimensional Stochastic Time Series Vectors

The project is aimed to establish and algorithm and computer software for generation of multiple time series of data which represent processes that are closely correlated and auto-correlated (i.e. data in time step i are correlated to the data in time steps i-1, i-2,i-3, etc.). This project is of significant importance to the company.  It represents significant improvement of our earlier developments in this field.

Implementing object tracking in head-mounted video capture for eyegaze tracking

Head-mounted eyegaze tracking allows experimenters to record the eye movements of a wearer while he interacts with a real situated environment. Analysis of the eye movements however, is difficult since motion of the wearer’s head causes objects to move relative to the head-mounted video camera. The focus of this project is to implement object recognition and motion tracking in video recorded from the head-mounted camera. Motion in the recorded environment will be matched to user’s eye movements to determine which particular object a person is looking at.

3D Visualization of the Electric Power Transmission Systems

The primary goal of the proposed internship project is to develop advanced 3D visualization methods and techniques for displaying transmission lines and other components of a power system in relation to the land cover and elevation maps. A software application will be developed using modern visualization approaches and proposed algorithms. The new system will be designed in such a way that it could be easily integrated with the current energy management systems (EMS/SCADA) and assets management databases, as well as with any other required software applications used by the project sponsor.

Design of a music retrieval system based on vocal queries

Currently users find their desired song on a multimedia sharing website such as YouTube using text based queries. The search phrase usually contains some text information about the song such as the name of the song, the singer, or parts of the lyrics. However, this information can be easily manipulated by malicious file up-loaders, who can use this scheme to upload pirated assets. This problem can be eliminated by letting users search based on the melody of the song rather than using textual information.

Motion Frameworks for Physical Expression

The intern will be porting and extending a tool for detecting and classifying qualities of human movement which is then subsequently fed to a realtime generative visualization system as part of an artistic process. The classification scheme is based on Laban Effort qualities. The system uses accelerometers and a neural network to recognize and differentiate qualities of movement by a performer.