Inspiration
Wanted to explore Geospatial domain, with the advancements in AI technology on an hourly basis it's important to keep up with the tech and understand various domains where AI is being implemented and how we can make use of it. Progressively understanding how well we can contribute to the tech. Hence we wanted to participate in this Hackathon and build product simultaneously learning from it.
What it does
It helps us determine if the FEMA reports generated are accurate and report any new findings and other validations
How we built it
Tech stack - AWS Sagemaker, S3 bucket , Twelve labs - Marengo and Pegasus, Python, VS Code, Streamlit
Challenges we ran into
Finding the dataset
Accomplishments that we're proud of
We worked on a track we never saw before and using techstack we weren't sure before. And in less than 20 hours to submit the hackathon we changed the problem statement but happy that we got good results. The time we incorporated to make this project is pretty meaningful and were able to get good results. It was meaningful.
What we learned
Learnt about Geospatial environment and new tech stack.
What's next for GeoSTL Track3-b
Enhancements - working with a streaming data, Locating the damaged areas on a map
Built With
- amazon-web-services
- python
- sagemaker
- twelvelabs
Log in or sign up for Devpost to join the conversation.