-
STUFFFIEEEES
-
Electronic Disposal
Inspiration
We thought that we should create a project to solve an environmental issue. Walking around made us realize that people actually don't sort their trash very well (strange observation I know). And so we decided as bored CS majors how much of an impact does improperly managed/sorted waste affect us socially, economically and environmentally, turns out many billions not only in monetarily but in terms of risk to wildlife and human health particularly with harsh chemicals etc.
What it does
To help solve the problem of improperly managed waste, we decided to make an intelligent robot that can identify commonly disposed items such as bottles, batteries, garments etc. and provide real time advice on recommendations to dispose or reuse certain products instead of just tossing them into whichever bin seems "right". Simply take a photo of your object in question and it will respond with some well advised advice.
How we built it
To construct this lovely advisor, we used Python with libraries such as, OpenCV, Google Gemini API/ Vertex API. Using an external camera or internal one such as those on a laptop or phone, we take a live video feed and when your object is in frame you hit spacebar and after a short delay you'll receive some lovely advice.
Challenges we ran into
The biggest challenge we ran into was accessing the google Gemini API, initially we had planned to use Java and the accompanying libraries, but found that there was little documentation on the Gemini API specifically for Java (or even python) due to it being such a new and evolving API, and that made the process much more difficult so we switched to Python. The drawback was python is much more resource intensive and slower at processing live video or even photos. Lastly it just happen to be that 2 of our team members were MacBook users which made terminal commands and collaboration about 79% more annoying. 💀💀💀💀
Accomplishments that we're proud of
Since this was our first Hackathon we didn't really think we would be able to come up with any kind of cohesive project but were really surprised and happy that not only did we actually create something but something that could potentially solve a problem implemented correctly.
What we learned
We learned a lot about how much effort and time it actually takes to develop something. Additionally working in a team environment and solving problems together is quite fulfilling and is different from working on projects solo. Aside from that we learned a lot about tools such as OpenCV, AI and API tools.
Log in or sign up for Devpost to join the conversation.