https://github.com/gangula-karthik/aicu-bike-search

Find Your Stolen Bike Lah! With AICU, We Kena Spot Your Bicycle on Carousell One Shot 🚲🔍💨
https://github.com/gangula-karthik/aicu-bike-search

chromadb clip computer-vision fastapi multi-modal natural-language-processing nextjs typescript

Last synced: about 2 months ago
JSON representation

Find Your Stolen Bike Lah! With AICU, We Kena Spot Your Bicycle on Carousell One Shot 🚲🔍💨

• Host: GitHub
• URL: https://github.com/gangula-karthik/aicu-bike-search
• Owner: gangula-karthik
• License: mit
• Created: 2023-09-29T16:29:23.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-09-16T07:41:27.000Z (3 months ago)
• Last Synced: 2024-09-16T09:16:55.105Z (3 months ago)
• Topics: chromadb, clip, computer-vision, fastapi, multi-modal, natural-language-processing, nextjs, typescript
• Language: Jupyter Notebook
• Homepage: https://www.youtube.com/watch?v=t4rZbfIGSLo
• Size: 28.5 MB
• Stars: 4
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


  

  


  Video Demo: https://www.youtube.com/watch?v=t4rZbfIGSLo



# AICU App

Welcome to AICU (Artificial Intelligence sees you) 🤖👁️🫵 - your digital detective in the quest to reunite you with your stolen bikes! Leveraging the cutting-edge technology of OpenAI's CLIP, AICU transforms the way you search for your stolen bicycles on online marketplaces like Carousell in Singapore.

## Features

- AI-Powered Search: Utilizes OpenAI's CLIP to create deep, meaningful embeddings from both text and images of bike listings on Carousell.

- Vector Database Integration: Incorporates embeddings into ChromaDB, ensuring efficient and accurate search capabilities.

- Fast & Accurate Results: Quickly browse through the most similar listings to find your bike with ease.

## How It Works

- Step 1: AICU generates CLIP embeddings for bike listings on Carousell, analyzing both images and text for comprehensive understanding.

- Step 2: These embeddings are stored in ChromaDB, optimizing search performance and accuracy.

- Step 3: Your query, be it an image or text, is then converted into an embedding and compared against the database.

- Step 4: AICU presents you with the listings most similar to your query, helping you find your stolen bike faster than ever.

## Star Our Repo!

Love AICU? Want to support our mission in combating bike theft? Here's how you can help:

- Visit Our GitHub Repository: Find us on GitHub under "AICUApp" and take a moment to explore our project.

- Star 🌟 the Repo: Show your support and help us gain visibility by starring our repository. It's a simple gesture that goes a long way.

- Spread the Word: Share our project with your friends, family, and fellow cyclists. The more people know about AICU, the stronger our community becomes.