Crowd Analytics

Crowd Analytics is a crowd counting and behavior analysis web app built with PyTorch and a Flask frontend.

Features

• Crowd counting from images
• Simple web UI for uploads and results
• Pretrained CSRNet model support

Tech Stack

• Python
• PyTorch
• Flask
• HTML/CSS/JS

Project Structure

• app.py - Flask app entry point
• model.py - CSRNet model definition
• img_inference.py - Inference utilities
• train.py - Training script
• templates/ - HTML templates
• static/ - Frontend assets
• files/ - Model weights and sample inputs

Setup

1. Create and activate a virtual environment

   python3 -m venv .venv
   source .venv/bin/activate

2. Install dependencies

   pip install -r requirements.txt

3. Download/ensure model weights

   • Place CSRNet weights at files/CSRNet_Modified.pth

Run

python app.py

Then open your browser at http://127.0.0.1:5000.

Demo

Demo Video

Notes

• static/uploads/ is generated at runtime and is ignored by Git.
• Large files (*.pth, *.mp4) are tracked with Git LFS.

License

MIT License. See LICENSE.

Contributors

• Anil Kumar
• Mounika
• Abhi Ram
• Manish