Inspiration
With recent events such as mass shootings and stabbings, there is a growing concern about safety in public places. An app that can detect flagged offenders and alert authorities and community members could be a valuable tool for increasing safety in schools, churches, malls, and other public spaces.
What it does
WatchDog is an app that uses security cameras and advanced machine learning technology to detect offenders in real-time. If a person is detected in multiple cameras set up within this technology, the app immediately sends notifications to authorities through a user-friendly web app, alerting them of the person's location. The app is designed to help keep communities safe by providing a quick and effective way to respond to potential threats.
How we built it
WatchDog was built using React and Flask. We also implemented Firebase for backend, and the machine learning library is based on YOLOv8 and a Re-ID Image Recognition Model trained on custom data.
Challenges we ran into
Integrating all the backend elements with the frontend elements in ReactJS involves combining the different parts of the application so that they can work together seamlessly.
Accomplishments that we're proud of
Learning how to work with ML Models. Our team also had to gloss over research papers to learn how things work and how can integrate them to our database.
What we learned
We learned how to conquer new tech stacks and better programming methodology.
What's next for WatchDog
If this project gains enough support, work on more R&D and establish a polished product, marketed towards authorities and security companies.
Log in or sign up for Devpost to join the conversation.