DripCheck

Inspiration

We wanted to create a platform combining social media interaction with fashion technology. The idea of DripCheck emerged from our love for expressing individuality through style and sharing it with friends, while also exploring how AI can enhance fashion insights.

What it does

DripCheck features a camera interface, similar to Snapchat, allowing users to capture moments. Swiping right brings users to the “Stories” section, where they can view photos shared by their friends. Swiping left leads to the “Gallery”, where users can browse their own saved photos. After taking a picture, DripCheck runs an AI-powered analysis of the outfit, providing instant feedback and insights about the fit, trends, and style breakdown.

How we built it

DripCheck was built using a combination of front-end, back-end, and AI Vision Processing interfaces. We developed the camera and gallery interfaces using Swift IOS Development, XCode. The backend infrastructure was powered by Swift as well and was created to be integrated with a machine-learning model built with TensorFlow for outfit analysis.

Challenges we ran into

One of the biggest challenges was training the machine learning model to accurately identify different clothing items and evaluate styles. We had to collect a substantial dataset and fine-tune our model to recognize various types of fashion trends. Although we made significant progress in training the Neural Network, we ultimately decided to pivot to using OpenAI vision processing. Given more time and resources, we would continue to build out machine learning to increase the specificity of the app and gear it towards more fashion-forward trends.

Accomplishments that we're proud of

We’re proud of building an app that combines fashion, technology, and social connectivity into a single platform. The successful integration of real-time AI analysis and a user-friendly interface is a major achievement for our team.

What we learned

Throughout this project, we deepened our understanding of machine learning model training and image recognition algorithms. We also gained valuable experience in optimizing app performance for real-time functionalities and improving user experience design to make navigation intuitive and enjoyable.

What's next for DripCheck

Looking forward, we plan to enhance DripCheck by introducing more advanced outfit recommendations and trend analysis. We aim to incorporate personalized style suggestions based on user preferences and previous analyses. Additionally, expanding the social features to include direct interactions, such as fashion polls and collaborative galleries, is on our roadmap. Further improvements to the AI model for greater accuracy and diversity in fashion recognition are also in the works.

GitHub Link

Built With

apiintegration
deepfashion
python
swift
tensorflow

Submitted to

HackNC 2024

Created by

Along with Aidan, I worked mostly on the machine learning and backend portion of the project. I got the opportunity to work on something new- Neural Networks with a processed dataset (DeepFashion)!

Neha B
I came up with the idea and had the vision for the app. I made the UI of the app and coded the pages on the front end. We all helped each other implementing the AI model to the app in order to generate feedback. I coordinated the process of designing the app in the beginning.

Saul Reyes
My contribution was working on the machine learning and backend portion of the project. Implementing OpenAI API to Swift files ensures a secure connection in the application. Another responsibility was working on the experimental Neural Network with a processed dataset (DeepFashion). Although I didn't have a Mac, I also coded the front-end and add styling to the front page on another member's working directory

aidanjlee5 Lee
I, along with Sven, designed our UI, and made Swift backend files for data storage of pictures and text. I also worked on connecting Aidan and Neha's backend to our frontend, and handled the setup and usage of our version control.

Advik Arora