Inspiration
Our inspiration for the Sorté project came from a strong desire to address the environmental challenges posed by improper waste disposal and the pressing need to promote recycling. We were motivated by the growing concern for the planet's well-being and recognized the potential to make a significant impact by creating an innovative solution for efficient recyclable storage and sorting.
What it does
Sorté is an intelligent recyclable storage model that streamlines the process of collecting, sorting, and managing recyclable materials. It utilizes cutting-edge technology such as computer vision and machine learning to automatically identify and categorize recyclables, making it easier for users to participate in recycling efforts. Users can simply deposit their recyclables into Sorté, and the model takes care of the rest, ensuring proper sorting and organization.
How we built it
We, in particular, used a Siamese network to detect similarities and distinguish between recyclables and trash. The Siamese network played a crucial role in our model's ability to recognize and categorize various materials accurately. This network was trained on a vast dataset of images of different pokémon, allowing Sorté to learn the subtle differences and similarities between different items.
Challenges we ran into
We initially explored the possibility of utilizing the COCO library, a comprehensive resource encompassing a wide array of common objects, including transportation, people, and materials. However, we encountered a significant hurdle as our dataset lacked labels. It became apparent that creating labels for the vast COCO dataset would entail a prohibitively time-consuming process. Consequently, we opted for an alternative approach by working with a more manageable dataset. In this new approach, we turned to a collection of diverse, random objects, and, intriguingly, this unconventional dataset was themed around Pokémon characters.
Accomplishments that we're proud of
We're immensely proud of several accomplishments with Sorté. First and foremost, we've created a tangible solution to promote recycling and reduce waste. We've successfully harnessed the power of artificial intelligence to automate and optimize the recycling process. In addition to that, we are extremely proud of the fact that we managed to build our Siamese network model in the span of a hackerthon and have it running.
What we learned
The development of Sorté has been a tremendous learning experience. We gained in-depth knowledge of computer vision and machine learning. We also developed a deeper understanding of the complexities of waste management and recycling systems, as well as the importance of aligning our project with sustainability goals. Our team also improved our collaboration and problem-solving skills.
What's next for Sorté
The future of Sorté looks promising. We plan to refine the model further, expanding its capabilities to recognize an even wider range of recyclable materials. Additionally, we aim to partner with municipalities, recycling centers, and waste management companies to integrate Sorté into existing systems. Ultimately, our vision is to see Sorté become a common fixture in households, businesses, and public spaces, contributing significantly to the global recycling and sustainability efforts. We are also exploring options for scaling up production and reducing the environmental footprint of Sorté itself to align with our sustainability goals.
Log in or sign up for Devpost to join the conversation.