A Collection of Exciting Projects
Praxis II Team Generation
Drone Computer Vision
It has become increasingly important to keep up to date with the developments in machine learning/deep learning technology in order to remain competitive in the field of computer vision.
Hence, we initiated a collaboration between UTAT related to developing a deep learning based object detection pipeline to be used on their aerial vehicles.
To this end, we have trained a model that is able to detect the precise location and class of target objects that enter the field of vision of the drone. We have also built a synthetic data generation system that can be used to generate realistic looking data with ground truth labels, which can be used to train the object detection system. We are looking to improve this system and augment it with further capabilities including color and rotation detection without compromising efficiency.
The U of T Course Clustering project aims to provide U of T students with more insight in different courses by grouping similar courses together using clustering algorithms. Throughout the past year, team members of this project got to explore different machine learning algorithms, strengthen their programming skills and have hands-on experience in web scraping.
The Team Builder Project is investigating how a class of first-year engineering students can be optimally placed into groups for school projects. If successful, the method generated here will be used to build teams for students in the Praxis II class.