Stylit

• 

  Area where you can see your like hairstyles

• 

  Distinguished gentleman using the photo feature

• 

  Profile options tab

Accomplishments that we're proud of

• Style-IT is a hairstyle matching application based on Multi-Layer Perception Neural Network Model which a Deep Learning Model for classification and further K-nearest neighbour Machine Learning Algorithms to find the best Recommendation for your hairstyle.
• We feature engineering to go beyond traditional image recognition and developed special features such as extra vector data / vector angles when computing the facial structure categorization,
• We take note of the distance, depth, and facial features to provide better accuracy in the AI's decision-making when it comes to matching facial structures.

What we learned

• All four of our devs had never used react native, but after this event we all feel equipped, and more confident than before in It, as well as in the use of JavaScript and external dependencies such as (Google) Firebase and REST APIs.
• For the backend and machine learning section of our project, the majority of our team had never developed or seen any machine learning models, with NONE of us using deep learning in the past. After the weekend, however, we managed to train our own model and learn a great deal about Deep Learning.

What's next for Stylit

• In terms of content and monetization: We are planning to add an E-commerce functionality on the app where people can book their own stylist depending upon their budget and location. We are also planning to promote short ASMR and creative videos like Tik tok by creators advertising them and engaging more people on our App. Alongside we are also planning to integrate fashion recommender functionality to make it a one-stop solution.
• In terms of functionality: With the time constraint we didn't manage to connect the front-end and back-end of the application. With more time, we are confident that our team can make this app more robust and market ready.

Built With

• api
• deep
• deep-learning
• express.js
• firebase
• javascript
• learning
• machine-learning
• mlp
• python