Skin Sync is an Android application that helps users decide what to wear for photoshoots based on their skin, lip, and eye colors.
By analyzing these features and the colors of a specific location, the app provides outfit recommendations to ensure the user looks their best in every photo.
It also scrapes clothing suggestions online, including prices and purchase links, making the experience seamless and efficient
In this built a system that can automatically generate quizzes and explanations from any uploaded PDF.
I used a Retrieval-Augmented Generation model, or RAG, but with a twist — I added a graph neural network to help the model understand complex relationships in the data.
This approach makes learning more interactive and helps with deeper understanding, especially for exam preparation.
An end-to-end deep-learning solution that automatically generates natural-language captions for any video by extracting visual features frame-by-frame and decoding them with an attention-based transformer model—making your content instantly accessible, searchable, and ready to share.