Jayleen Zhou

Tell us a bit about yourself before you enrolled in the Visual Computing program at SFU.

I learnt general computer science in my undergrad studies and worked as a front-end developer in digital agencies for the past three years. I worked on digital campaigns, landing pages for all sorts of clients and creative promotion web application.

I am interested in applying technology into digital content creation. One day, I was preparing a pitch for a client who sold pet accessories. I researched creative campaign examples and found a video demo of an augmented reality installation that Purina did in a shopping mall. Purina was marketing its cartoon mascot ‘Felix the cat’. When a customer stood in the specified area in front of Purina’s camera, they showed up on the big display of the mall, together with Felix as a cartoon character within the same scene. Felix could respond to the customer’s action and interact with the surrounding objects. Watching the video, I start to truly sense the attractiveness of augmented reality. I believe adding more of this kind of installations to our daily life will change the way we live, and I want to be one of those people who can create such content.

What attracted you to the visual computing program? Why did you choose to come to Canada?

Vancouver is a well-known hub for visual effects and digital animations, and SFU has a good reputation on visual computing on the programming and research sides. Therefore, I thought taking this freshly-launched Master program can help me learn 3D programming and can jump start my career in the field.

What is the most important thing you have learned in the program so far?

Before diving into mathematical steps or learning algorithms, I imagined what deep-learning based computer vision might look like, but I only had a vague idea. In one of the courses last semester, we walked through the concepts from convolution, to feature maps, to neural network models and then were introduced to a currently well-accepted object detection method - Single Slot detection. We also read the original paper and recreated the key algorithms from scratch using PyTorch. Because of the step-by-step learning, I do not find the term “object detection” vague any more. This experience also reminds me that if I need to learn something in the future, I need to break it into pieces and tackle each section one-by-one.

Why do you think this program is important or significant?

The program covers lots of aspects related to visual computing such as object detection, image synthesis, image compression, 3D modelling, 2D SLAM and 3D SLAM. We have hand-on practice in each of the topics. The professors provide detailed lectures to help us to learn well-known concepts. The programming labs reinforce this learning, so that after we go through all these steps, we are able to carry on our understanding and apply the concept to real-life problems.

What types of jobs are you looking for? How does this program fit with your career goals?

I am looking for content creation jobs after graduation. Taking computer vision courses, I now understand better and deeper about the concepts behind the applications I have used before. For example, the feature tracking and the visual SLAM algorithms has helped me understand the foundations of locating and tracking an object in motion in 3D space, which is part of the underneath algorithm of WebAR.js. The computer graphics course that I am in this semester explains 3D modelling concepts, which are implemented in three.js library.

What would you tell potential students looking at SFU’s Visual Computing program at SFU?

The program is intensively mathematical. Before coming to the program, it will be better for incoming students to refresh their knowledge of statistics and linear algebra before the semester starts.