Inspiration
Growing up in rough areas, we thought that crime and weapons crime in particular was something we wanted to address. We wanted to make the communities safer and people live without always having the feeling of fear crawling behind them.
What it does
Mobile webapp which allows users to report incident discreetly, detect knives in media feed, and trace nearby stabbings on maps. To reduce panic, we sent the incidents to authorities to let them flag them as important or not.
How we built it
Detectronv2 (Facebook object detection) trained on our own dataset + fine tuned. React native + expo mobile webapp. React webpage. Google Maps API for incident tracking
Challenges we ran into
Developing our own ML knife detection. Sending media through frontend to backend.
Accomplishments that we're proud of
Managing to get the knife detection successfully working on our own dataset. Building a full stack mobile webapp.
What we learned
How to train ML models. Build mobile webapps. Use Google Maps API.
What's next for SHANK PROOF PRO
Scrape CCTV and just feed it to our knife detection to automatically detect knives in public.
Built With
- detectron
- expo.io
- react
- typescript
Log in or sign up for Devpost to join the conversation.