Style Swipe

StyleSwipe merges dating app swiping with fashion discovery, allowing users to instantly purchase outfits they like, streamlining shopping.

Comment

Inspiration

We wanted to make a project that would allow users to get fashion inspiration and information based on clothes they like, which they could demonstrate through an interactive like and dislike button. It’s like a dating app, but instead of getting information about people, you get information about clothes you like.

What it does

StyleSwipe revolutionizes the online shopping experience by melding the addictive swiping features of dating apps with fashion discovery Users are presented with a tailored selection of outfits, and with a simple swipe right, they express interest in an ensemble What sets StyleSwipe apart is its seamless integration of e-commerce functionality: upon indicating interest in an outfit, users are promptly provided with direct links to purchase the individual clothing items comprising the ensemble, streamlining the shopping process This fusion of intuitive design and effortless purchasing transforms browsing into a dynamic and efficient experience, making StyleSwipe an indispensable tool for fashion enthusiasts seeking to curate their wardrobe with ease and style

How we built it

We first created the website design on Figma and then used traditional HTML and CSS to create the website. In addition to that, we integrated CloudFare AI to create a personal stylist to help you discover your color palette and outfit style. The backend features include BeautifulSoup and Selenium used for Web Scraping to get the outfit images.

Challenges we ran into

We also ran into some difficulties while scraping image links off of Pinterest, we had to ensure that the images were in fact from our Pinterest Board and that the links led to the correct images. We initially used BeautifulSoup, but had to pivot to Selenium because of Pinterest’s dynamic loading nature. BeautifulSoup led to missing extractions with images and links. Initially, we were going to use Nextjs as the web framework but there was an error with returning the response from the model, so we switched to Flask also defeating the need to find a way to integrate Python with a jsx framework We also ran into an issue of figuring out how to deploy the website application as we used Flask for the first time.

Accomplishments that we're proud of

We wanted to make sure that the website looked appealing while also giving the user information that would be useful to their experience, which we think we were able to do.

What we learned

We learned how to implement APIs and create an interactive website using tools like Figma and how to implement our design using Flask.

What's next for Style Swipe

We could implement more advanced image recognition algorithms with ResNet by Cloudflare. We could also implement an algorithm that recommends images based on brands that the user frequently likes or interacts with. This could expand to include accessories. Another algorithm could also build a user’s color palette based on their skin color, hair color, and eye color. We want to possibly even avoid showing data similar to the photos that were disliked, which would most likely require another API or libraries like Surprise from Python.

Share this project:

Updates