I am currently a PhD candidate at the University of Toronto Institute for Aerospace Studies (UTIAS). I work under the supervision of Professor Steven Waslander as a part of the Toronto Robotics and Artificial Intelligence Laboratory(TRAIL). I specialize in perception algorithms for autonomous driving. Specifically, I work on fundamental research within 2D and 3D object detection, semantic segmentation, uncertainty estimation for deep neural networks, and synthetic data generation.
Recent News
November 16, 2019
Code for BayesOD released.
Released code for our probabilistic object detector: BayesOD. The code can be found here.
November 08, 2019
Talk at the IROS 2019.
Presented a talk on how to evaluate probabilistic object detectors in IROS 2019 workshop on The Importance of Uncertainty in Deep Learning for Robotics.
August 01, 2018
Transfering To The University Of Toronto
Starting September 2018, I will officially be a Full Time PhD student at UTIAS (University of Toronto Institute for Aerospace Studies). I am no longer associated with the University of Waterloo.
June 13, 2018
Talk at Vector Institute Endless Summer School
I will be speaking at the Vector Institute Endless Summer School. My talk targets the limitations of the current state of the art object detectors and future research directions in that area.
MY LATEST RESEARCH
The results of our probabilistic object detector, BayesOD on the KITTI 2D object detection dataset. BayesOD provides reliable estimates of uncertainty for the category and the bounding box of all objects in the scene.
The results of the 3D object detector me and my team designed. The detector runs at 13 Hz with a 2GB memory foot print making it an optimal candidate to be deployed on autonomous vehicles.