VFind is a web application that allows users to search for products using images. Upload an image to quickly find similar products online. The app also features a virtual try-on capability powered by Google's Gemini AI.
- Next.js, Tailwind CSS, shadcn/ui, Framer Motion
- Python (FastAPI for backend), Pinecone (img to vector, vector search)
- MongoDB (for storing product data)
- PostHog (for web analytics)
- Google Gemini 2.0 Flash (for virtual try-on feature)
- Visual Search: Upload product images to find similar wearable items
- Virtual Try-On: Visualize how clothing items would look on a person using Google's Gemini AI model
- User Authentication: Personalized experience with user accounts
- Responsive Design: Works on all devices
- add signup and login functionality
- backend API isnt secure, API endpoint can be retreived via dev tools (anyone could use API directly), api route must be implemented
- implement virtual try-on feature using Google Gemini 2.0 Flash
- 10 similar results or something like that (if user is signedup)
- full body fit recommendation (if user is signedup)
- improve virtual try-on accuracy and performance
- browser extention (if user is signedup)(only top 3 similarsearch result)