This project is an Image Detection web app built with Hugging Face Transformers and Gradio. It allows users to upload an image and get predictions for its content using a pretrained vision model.
Try the live app on Hugging Face Spaces:
https://huggingface.co/spaces/vysubs28/image-detect-app
- Upload images easily via web UI
- Uses state-of-the-art vision transformer model for accurate classification
- Clean and responsive interface built with Gradio
- Runs on CPU/GPU or Apple Silicon MPS if available
To run locally:
git clone https://github.com/yourusername/image-detect-app.git
cd image-detect-app
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt
python app.py
Usage
Open the app
Upload an image
View the predicted label and confidence score
Technologies Used
Python 3.11
Transformers (Hugging Face)
Gradio
PyTorch
License
MIT License