We are opening our doors to talented students with expertise in mathematics, signal processing, electrical engineering, computer science and design. Students with a strong grounding in theory as well as great enthusiasm for hands on hacking would fit in well with the group. Before submitting an application, take a chance to look through the research and the publications to decide on the work that most fits in with you. If you would like to a learn more about this field, it's a great idea to purchase the Intelligent Image Processing and also take the ECE 516 course available to University of Toronto students. Feel free to contact any of the people in our lab for more information on the project and how you can contribute. Please have a resume ready to show some past work and accomplishments.
The ePI Lab has had a successful history with undergraduates research students. Often, undergraduates have succesfully published their research work in internationally prestigious journals. For those interested in graduate work or employment, publishing can be an excellent experience.
|
EyeTap Prototype DesignThe EyeTap is the key enabling technology for mediated reality. We are currently seeking a talented group of individuals with a strong design and engineering background who would be capable of handling the cross-disciplinary issues involved in designing and making EyeTaps. Ideal candidates would have a strong ability to build things, yet also understand issues of optics and electronics. For specific information, contact the project heads: Chris Aimone, and Steve Mann |
|
libD2HThe libD2H allows for direct control of cameras using the ptp language. This allows you to directly hack a camera and all its functionality through a computer system. Currently, we can control many of the parameters of a Nikon D2H through either a powerbook Mac or linux system. This remote control of the Nikon camera is useful for many powerful applications. If you are interested in learning about how digital cameras work inside, and having the ability to hack them, then this is a fun project for you. To learn about the specifics, contact the project heads: Corey Manders. |
|
OpenVIDIAThe world's first GPU based Computer Vision and Mediated Reality Processing library was created by the ePI Lab. Graphics Processing Units (GPU) have previously been used for rendering graphics, but they have never been used to process images. The GPU, with its ability to process matrices very quickly and effectively, can outperform many CPU in matrix specific tasks such as image and signal processing. This is an excellent project for students interested in learning more about computer architechtures, while also picking up high demand skills such as cg which are constantly needed by the graphics and gaming industries. For specific information, contact the project head James Fung. |
 |
Med-R ToolkitThe mediated reality tool kit implements a revolutionary technology which allows for 3D and 2D virtual content to be seamlessly and realistically introduced into a scene without the need of markers. Much of the new algorithms are based around the personal imaging technologies of video orbits and lightspace, which have recently been introduced into the OpenVIDIA library. To learn about the specifics, contact the project heads: Billal Belmellat, Mohit Kansal, and Andrew Tae-Hyung Kwon. |
 |
JPEG Lightspace Map(JLM)The JLM is the future imaging standard that has been considered by an major electronics manufacturer and will be important towards achieving high tonal fidelity towards the realization of analogue or "undigital" images. Although we have implemented the JLM library in C, there remains many interesting topics including optimization of compression to acheive this standard. Also, there are exciting topics including the implmentation of JLM compression on commercially available cameras through the libD2H library. To learn about the specifics, contact the project heads: Billal Belmellat, and Mohit Kansal |
