Skip to content
Nymna Alert - Anomaly Detector

Nymna Alert - Anomaly Detector

Enhancing Safety with ML-Driven Surveillance.

Created on 11th May 2024

Nymna Alert - Anomaly Detector

Nymna Alert - Anomaly Detector

Enhancing Safety with ML-Driven Surveillance.

The problem Nymna Alert - Anomaly Detector solves

Minimizes the response time with real time monitoring and instant alert system thorugh mobile app.

Challenges we ran into

Since this was our first hackathon, the main challenge we faced was the distribution of tasks among each team members.
Moreover, we had some challenge with figuring out the time to perform the delegated tasks , which ultimately would make us awake for the whole 2 days.

Regarding technological challenge, the main issue was training the Computer Vision model due to the lack of resources and data available to us. It was hard for us to maintain a high level of accuracy due to this lack of resource. We had underestimated the amount of data we would require to train the model, so a whole night of work was wasted due to lack of annotated data. Moreover, data annotation also took a long time to prepare.

While we were looking for a greater scoped anomaly detection that would cover fights, violence, weapon detection and more, we could only do little due to this little time frame, but if time was on our favour, the possibilities were endless.

Despite all this, the experience was incredible and we had a thrilling experience.

Discussion

Builders also viewed

See more projects on Devfolio