Beer Detection with YOLO

This project performs live beer detection using a YOLO model through your webcam.

Files

• detect_beer.py - Live webcam detection script with buffering
• yolo_beer_detection_tutorial.ipynb - Educational notebook for training and understanding YOLO
• pyproject.toml - Project dependencies
• best.pt - Your trained YOLO model (place here)

Quick Start

1. Place your YOLO model file (best.pt) in the project root directory.
2. Install dependencies: uv sync
3. Run the detection: python detect_beer.py

Educational Notebook

For a complete learning experience, open yolo_beer_detection_tutorial.ipynb which covers:

• YOLO fundamentals
• Data preparation
• Model training
• Evaluation
• Inference and deployment

Requirements

• Python 3.8+
• Webcam
• YOLO model trained on alcohol-related objects

Usage

Run the script and point your webcam at beer bottles, glasses, etc. The model will detect and highlight:

• alcohol-bottle
• beer-bottle
• beer-glass
• shot
• wine-glass

Press 'q' to quit the detection window.

Model

Make sure your best.pt model is compatible with Ultralytics YOLO26 format.