I'm currently a first year MSc student in Computer Science (Machine Learning Group) at the University of Toronto since September 2016. I am supervised by Raquel Urtasun. Prior to that I completed my BASc in Mechanical Engineering (Mechatronics Program) at the University of British Columbia (2011-2016). My research interests lie on the intersection of Computer Vision and Machine Learning. My CV is available here.
My e-mail is justinliang at cs dot toronto dot edu.
CSC2541: Topics in Machine Learning - Sport Analytics (Winter 2017)[PDF]
Gave a lecture on two activity recognition papers: End-to-end Learning of Action Detection from Frame Glimpses in Videos and Detecting Events and Key Actors in Multi-Person Videos.
Unsupervised Learning of Basketball Offensive Plays
Implemented a combination of a CNN and LSTM to generate offensive basketball plays given the basketball and defensive basketball player trajectories.
Detection and Tracking of Basketball Players
Performed detection and tracking of basketball players using Faster-RCNN and Deep SORT.
Egg Yolk Consistency Measurement Device
Designed and developed a device that measures the yolk consistancy of an egg by oscillating it on a spring.
Elevator Simulation [CODE]
Simulated an elevator system using principles of concurrent systems such as data pools, pipelines and semaphores in C++.
Trajectory Generation and Control for a Linear Motor Driven XY Table
Designed an integrator plus lead/lag controller and a pole placement controller to draw on a XY table. Developed functions to perform interpolations to draw a custom toolpath.
Motor Control Using a MSP430
Designed a proportional controller and used it to control a DC motor with a MSP430 microcontroller. This was done by taking in encoder data and adjusting the timer to produce the correct PWM duty cycle.
8 Bit CPU
Built a simple CPU in VHDL consisting of an ALU and a register file on the Altera Cyclone II FPGA.