Any2YOLO

Easily convert AnyLabeling JSON into YOLO txt files!

This tool simplifies dataset preparation for YOLO models, whether you're processing a single JSON file or managing batch conversions with multiple labels

---

🌟 Key Features

• Batch Conversion: Upload and process multiple JSON
• Label Selection: Choose specific labels for YOLO txt format or include all
• User-Friendly Interface: Intuitive GUI
• Detailed Logs: Monitor every step of the conversion process
• Error Handling: Skips invalid files and provides detailed error logs

---

🚀 Getting Started

Prerequisites

• Python 3.8 or newer

📦 Installation

Clone the Repository

To get started, clone the repository to your local machine:

# Clone the repository
git clone https://github.com/danilobirbiglia/Any2YOLO.git

# Navigate into the project directory
cd Any2YOLO

Install Dependencies

For Linux/macOS

Note: The tool is not fully optimized for Linux/macOS. Some features may have limited functionality

Install tkinter and Pillow:

sudo apt-get install python3-tk -y
pip install Pillow

For Windows

• Ensure Python is installed with the tcl/tk and IDLE option enabled during the installation process
• Then, install Pillow using:

  pip install Pillow

Verify tkinter Installation (Optional)

To confirm that tkinter is properly installed, run:

python -m tkinter

If a small GUI window appears, tkinter is correctly installed

---

Launching the App

1. Install dependencies:

   # Install Python dependencies
   pip install Pillow

2. Run the application:

   python src/main.py

3. Follow the on-screen instructions to upload and convert your JSON files

---

🖥️ How It Works

1. Run the App: Start the GUI with python src/main.py
2. Upload Files: Click the "📂 Upload JSON Files" button to select one or more files
3. Review Labels: View and select labels extracted from your JSON files
4. Convert Files: Click "Convert" to generate YOLO compatible .txt files
5. Access Results: Converted files are saved in the same directory as the input JSON files

---

📂 Project Structure

Folder/File	Description
src/main.py	Entry point for the application
src/gui.py	GUI logic for file management and conversion controls
src/converter.py	Core logic for JSON to YOLO conversion
src/logger.py	Manages logging for debugging and troubleshooting
src/styles.py	Defines styles for the GUI application
assets/any2yolo.ico	Icon for the GUI
logs/converter.log	Directory for log
README.md	Project documentation (you’re reading it!)

---

💻 Example

Input JSON

{
  "shapes": [
    {
      "label": "dog",
      "points": [[100, 200], [300, 400]],
      "shape_type": "polygon"
    },
    {
      "label": "cat",
      "points": [[50, 50], [150, 150]],
      "shape_type": "polygon"
    }
  ],
  "imageWidth": 500,
  "imageHeight": 500
}

Generated YOLO TXT

0 0.4 0.6 0.4 0.4
1 0.2 0.2 0.2 0.2

Steps Taken:

1. Extracted labels: dog, cat
2. Calculated bounding box dimensions and coordinates
3. Converted to YOLO txt

---

📜 Logs

Detailed logs are available through the "Logs" menu in the app

Example Log Output:

INFO: Initialized with files: ['annotations.json']
INFO: Extracted labels: ['dog', 'cat']
INFO: Successfully converted annotations.json to annotations.txt
WARNING: Skipped invalid shape in annotations.json

---

❓ FAQ

What happens if a file is invalid?

The app skips invalid files and logs the issue for review

Can I use this for bounding boxes instead of polygons?

Yes! YOLO txt files support both polygons and bounding boxes

---

📜 License

This project is licensed under the MIT License

---

👤 About the Creator

• Created by: Danilo Birbiglia
• LinkedIn: Danilo Birbiglia
• Email: danilobirbiglia@gmail.com

---