YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)
🧠 Fake Product Analyzer
An AI-powered system that detects product logos from images using YOLOv8 and classifies them as Real or Fake using ResNet50. Built as part of my AI/ML project during the CDAC Training Program.
🚧 Status: This project is still in development. More features and improvements coming soon!
🔍 Problem Statement
Counterfeit products cause major financial and health-related issues. This project aims to:
- Detect logos from product images
- Classify the logo as Real or Fake
- Provide explanation or reasoning using an LLM
- Offer an easy-to-use Streamlit interface
🚀 Tech Stack Used
| Component | Technology |
|---|---|
| Object Detection | YOLOv8 |
| Image Classifier | ResNet50 |
| LLM Integration | OpenAI |
| Frontend | Streamlit |
| Language | Python |
🖼 Sample Output
| Input Image | Detected Logo | Classification | Explanation |
|---|---|---|---|
 |
Nestle | Fake | The logo font doesn't match official Nestle branding. |
🛠 How to Run Locally
- Clone the repo:
git clone https://github.com/yourusername/Fake-Product-Analyzer.git
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support