At the VIP Lab Matthew has worked to improve the robustness of Visual Place Recognition to environmental effects through an MASc supervised by Dr. John Zelek. VPR is used to detect when a camera returns to a previously-visited location and is core to Simultaneous Localization and Mapping (SLAM), itself a mainstay of mobile robotics including self-driving vehicles. Matthew’s research has been focused on integrating visually-recovered 3D structure into VPR, as this is robust to illumination and viewpoint changes that hinder the versatility of appearance-based methods. Matthew holds a prior BEng degree in computer engineering from Toronto Metropolitan University with added focus in computer vision, robotics, and machine learning.

Email: m7bradle@uwaterloo.ca
Github: https://github.com/m7bradle