It is hard to keep up with the latest fashion trends and time-consuming to shop from store to store just to find a perfect fit of clothing. That’s why we build Project Mirror, providing AI-inspired advice to assist users in their fashion shopping experience.
What it does
When users run the client-side camera program and access the website, they will be given a view of themselves on the screen wearing that particular item of clothing. They will be given trials of various clothing and asked to swipe left or right based on their preferences. After enough data is collected, the user will be given a detailed report on what clothing is suitable for them based on their likes, current trends, and much more!
How we built it
The front-end was built with HTML and CSS using the bootstrap framework. The back-end was developed with Flask and the machine learning algorithm was developed with various python frameworks. The client-side camera program was made in python, using OpenCV and the media-pipe framework to get the key points in the body. The image of the clothing item is then overlaid on top of it using mathematics for precise placement.
Challenges we ran into
One major challenge we ran into was the integration of the front-end and the back-end. Alone, they worked perfectly fine but once assembled, they immediately broke down.
Accomplishments that we're proud of
We are proud of our smooth user experience and our deep analytical insights from our AI-inspired algorithms. The visualization technique used with CV was a huge accomplishment as well.
What we learned
The front-end engineer learned a lot about the usage of the bootstrap framework. The back-end team learned about the complexities of machine learning algorithms.
What's next for Project Mirror
Project mirror has the potential to impact millions. Our future steps include developing a mobile app that millions of users worldwide can access easily.