WALDO-Enhanced-Aerial-Detection

Overview

This repository builds upon the WALDO object detection AI model to offer enhanced capabilities for analyzing aerial footage. In addition to the object annotations provided by WALDO, this project also generates an enhanced 2D map to give a bird's-eye perspective of the detected objects. This enables a transformative viewpoint that can serve multiple use-cases including urban planning, real-time traffic monitoring, and aerial surveillance.

Features

• Real-time object detection and annotation in aerial footage.
• Enhanced 2D mapping with icons color-coded to match WALDO's detection labels.
• Output video saving functionality.

Requirements

• Python 3.x
• OpenCV
• ONNX Runtime
• NumPy
• Other dependencies are listed in requirements.txt

Setup

1. Clone this repository.

2. Install the dependencies:

   pip install -r requirements.txt

3. Place your aerial footage videos in the ./input_vids folder.

4. Copy the desired model to the playground folder.

5. Add the main.py script in the playground folder.

6. Run the main script:

   python3 main.py

Acknowledgments

This project leverages the powerful WALDO object detection model. A huge shoutout to the original creators of WALDO for their exceptional work.

License

This project is open-source and available under the MIT License.