Detection of Polluting Plumes ejected from NYC buildings

Identified smoke plumes emitted from buildings in NYC using a building-top camera images, classified using deep- learning based object detection. Developed a web app for image streaming, on-the-fly processing, and tagging.

We worked on detecting, counting, and tracking instances of plumes ejected from NYC buildings to help gain a better understanding of the building metabolism and the environmental and health impacts of building energy use. We received recognition for the project and won 1st place among all NYU CUSP Capstone projects. Check out our project website for more information!

Also, we created tools for building image pipelines and applying labels to images that are organized in a consecutive manner. Go check out our repo for more information on how to install and use it!

Our Project Website! Our Repo