About the SFU Vision and Media Lab

The SFU Vision and Media Laboratory conducts research in computer vision and multimedia. We develop algorithms for interpreting and processing visual data — images and videos. These algorithms facilitate a range of applications from enabling video search to enhancing the visual quality of photographs. We focus on image and video analysis, object recognition, human activity recognition, and machine learning.

Human activity recognition: Understanding what humans are doing in images/videos is a core thrust of our research. The grand goal of the field of human activity recognition is to build systems that can find humans in either still images or video sequences, and determine what action they are performing. Applications include internet video search, security, surveillance, or sports analytics. Our previous work has spanned this gamut, ranging from analyzing team sports video footage, to YouTube video tagging, to processing surveillance video from nursing homes to understand and prevent falls by elderly residents.

Object recognition: Automatically identifying and describing objects in images can be done with computer vision algorithms. Naming the objects in an image (e.g. this picture contains a fridge, a microwave oven, a sink), describing the high-level scene (e.g. this is a picture of a kitchen), or describing properties (this object is round, shiny, and inedible) are problems addressed by our research. The approaches we develop emphasize machine learning algorithms trained to utilize structures and relations between objects in a scene.

Lighting Invariance: Humans are adept at handling changing illumination: we tend to ignore the effects of shading and shadows, and we use bright highlights as a useful visual cue, rather than a source of confusion. But computer programs aiming at image understanding see these factors as distractions. One strand of research in the VML looks at removing these distracting factors using image processing to produce images that are invariant to lighting change. One compelling outcome of this work has been the development of images that are unchanged from the original, except that the shadows are removed.

For Prospective Students and Potential Industry Partners

We are always looking for motivated, good graduate students. We are also keen on fostering links with industry for joint projects that are based on research grants and/or student internships that allow graduate students to spend time at companies doing high-risk high-reward research projects. We are grateful to have received funding from and conducted joint research projects with a host of international and local companies including Google, Disney Research, NVIDIA, MDA Corporation, Sportlogiq, and many others.

Please contact any of the following faculty members:

If you are a graduate student in the School of Computing Science and would like to know more about the lab and our activities, or if you are a prospective graduate student, please contact us for more information. We are looking for qualified and motivated PhD students. Please note however, if you are applying for graduate school to the School of Computing Science at SFU, you are admitted only on the merit of your academic credentials and not based on your potential interest in any particular research lab.


The SFU Vision and Media Lab and the faculty offices are located on the SFU Burnaby campus in the TASC-1 building (Technology and Science Complex 1), just south of Science Road and north of South Campus Road, east of the South Science Building. The lab is in Room 8000.

If you come by bus, stay on until the last stop on campus (the Bus Loop next to Lot E). It is a short walk south to TASC-1. If you drive up, you will have to find parking in one of the visitor spots and make your way to the TASC-1 building. The closest visitor parking is Lot VB.

Mailing Address

Vision and Media Lab
Simon Fraser University
Technology and Applied Sciences Building 1: TASC-1 8000
8888 University Drive
Burnaby, B.C. V5A 1S6