Inspiration
As a team we all identified climate change as one of the biggest problems in the world that needed to be addressed. We were inspired by Inrix's idea of measuring traffic data and decided to base our application on traffic image data.
What it does
Our project uses traffic image data from the Santa Clara Traffic Cameras and takes advantage of computer vision to identify the number of cars on the road.
How we built it
We took advantage of AWS's Rekognition computer vision software to identify and count the number of cars on the road. We also used flask to host the app altogether as well as to fetch and consolidate the image data.
Challenges we ran into
The main initial challenge we faced was finding a unique way to apply AI without reinventing the wheel. We tried to think of problems in our daily lives that we thought AI could help with but a quick Google search always seemed to reveal an existing AI application for each use case we thought of. It was Juan's idea to think of potential ideas in terms of different emotions, and as a team we all agreed that climate change was one of the biggest issues that AI had the potential to help with.
Accomplishments that we're proud of
While some of us had some previous exposure to AI and ML, none of us knew much about using AWS and its many features. Also, none of us (besides Billy) knew much about flask. There was an initial learning curve with trying to immerse ourselves in all this software in such a small timeframe but even then we feel surprisingly well acquainted with AWS after barely even 24 hours.
What we learned
The biggest thing we learned about was all the different features of AWS, and how they could be used to build AI powered applications and projects. Even though we opted not to use it, we were in awe at how powerful Bedrock was and we could instantly see how it could be used to easily build and train chatbot agents for various use cases. We also learned how to use flask to implement the web and API aspect of out project.
What's next for Rekognizing Climate Change
The next step would probably be to consolidate all our insights into a service that could be monetized at scale. We all anticipate that AWS's cloud services would be extremely useful for this.
Log in or sign up for Devpost to join the conversation.