This repository is part of the delivered solution I researched and developed for Triple. It was executed as part of my bachelor's thesis @ University of Applied Sciences Leiden.
The goal of this project was to investigate the feasability, attractiveness and possibilities of performing real-time computer vision inference in a scalable cloud environment through Azure. The results propose a solution that is able to utilize the cloud's high scalability to deliver a highly customizable automated platform that takes the responsibility of analyzing livestreams for all sorts of different use cases. In an optimal scenario it can execute the process of receiving frames, analyzing them and publishing them within a second and below a cost of €1/hour.
