The fashion industry is the second biggest polluting industry in the world, behind transportation. Today, we consume 20 times more clothing than just a decade ago yet we wear them for half as long. This problem is worsened by the increasing popularity of online retail stores, particularly hyper-fast fashion brands such as Shein or Yesstyle, whose popularity has grown exponentially in the past few years. “Fast fashion” is the production of cheap and trendy clothing which is often discarded when its popularity dies or when its material expires. This forms a very unsustainable cycle of overproduction and consumption. The use of cheap textiles like polyester (derived from fossil fuels), the exploitation of workers who are forced to work in dangerous environments, and the fast-fashion culture in which cheap clothing is bought and discarded almost immediately, all contribute to the very toxic, unsustainable, and unethical fashion industry. We wanted to find a solution to this, and so thought of susFashion.
What it does
susFashion provides eco-friendly alternatives to any piece of clothing a user desires. With the upload of any image, susFashion will search sustainable online storefronts for similar items, and provide basic information such as name and price. Online consignment stores such as ThredUp or theRealReal provide a vast selection of gently used clothing items, sourced with user donations and donations from brick-and-mortar consignment stores. It is a simple and efficient way for someone who wants to do their part in supporting the environment to do so while simultaneously catering to their personal taste.
How we built it
Challenges we ran into
One challenge we ran into from the start was figuring out how to get our program to take an image and find out what clothing item it was displaying, and then search the internet for sustainable options. We decided to look for an API but did not find anything suitable that was also free. We settled with LykDat, a computer vision API that recognized clothing items. However, the issue with this API was that it was inaccurate and very unsure about its results. Our solution was to find the most common occurring words within the API’s results to find a common clothing theme. This algorithm allows us to web-scrape clothing that are more closely related to the clothing item the user submits
Another big challenge we ran into would be connecting the backend to the frontend. Since we were working on the parts separately, we assumed the connections would be the simplest. However, when it came to actually connecting the backend and frontend, we realized transferring files through an API would be difficult. We considered using a database, but that would have taken up a lot of our time. We settled with simplifying the front end by requesting an image URL from the user rather than an entire file. This not only saved us a lot of time, but also let us get the basic functionality of the program to work.
Accomplishments that we're proud of
Overall we are proud that we were able to put together a functioning program within the time constraints, having people with little to no experience with hackathons on our team.
What we learned
We learned a lot about the basics of web development both on the frontend and backend. We also gained a lot by creating our own API and connecting that to the frontend.
What's next for susFashion
One big thing that would be next for susFashion would be improving the UI/UX. At the moment, it’s a very simple webpage that searches for sustainable options based on a link to an image. In the future, we would like to make the page more dynamic and improve graphics past simple one colored elements. In addition, we would also expand our search of clothing beyond online consignment stores to physical consignment stores, slow fashion brands, or social commerce apps such as Poshmark. We would also complete the other sections of the page (About and Contact Us). Furthermore, we would like to develop it into a mobile app so that instead of importing images, users can take a picture of any clothing item from their phone and upload that directly onto the app. Finally, it would help to add filters based on price and location so that users can quickly find an item they want.