Skip to content

README.md

Latest commit

 

History

History
98 lines (63 loc) · 2.58 KB

README.md

File metadata and controls

98 lines (63 loc) · 2.58 KB

TagiFLY v1 (Beta)

Labelling Tool for AI Computer Vision

Open source, lightweight, and focused only on what you need.

📸 Screenshots

Main Screen Annotation Example
Main Annotation
Export Window Dark Mode Light Mode
Export Dark Light

🚀 What is TagiFLY?

TagiFLY is an open-source labelling tool for AI computer vision projects.
The goal is simple: Mark your images, save them, and use them in your AI/ML projects.

  • 🤖 Training data for machine learning
  • 🎯 Computer vision projects
  • 📊 Research & personal experiments

✨ Features

  • 📦 6 annotation tools: Box, polygon, point, line, mask, keypoints
  • 📤 4 export formats: JSON, YOLO, COCO, Pascal VOC
  • 🎨 Light / dark theme
  • ⌨️ Keyboard shortcuts for faster workflow
  • 🖼️ Supports JPG, PNG, BMP, GIF, WebP

📤 Export Options

  • JSON → Full project backup
  • YOLO → For object detection models
  • COCO → Research standard format
  • Pascal VOC → XML-based compatibility

⚠️ Note (Beta Release)

This is the first release (Beta).
It may contain bugs or missing features. Please share your feedback so we can improve it together. 🚀

🛠️ Installation & Setup

Prerequisites

  • Node.js (v14 or higher)
  • npm or yarn
# Clone the repository
git clone https://github.com/dvtlab/tagiFLY.git

# Navigate to project directory
cd tagiFLY

# Install dependencies
npm install

# Start the application
npm start

🚀 For Non-Technical Users

Download the ready-to-use application:

⚠️ Note: These are beta builds.
If you encounter any issues or have suggestions, you are welcome to provide feedback.

🎮 Usage

  • Select an image folder
  • Pick a tool and start annotating
  • Export your work in JSON / YOLO / COCO / VOC format

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

Open Source & Free - Use it anywhere, modify it, contribute to it! 🚀

⭐ If TagiFLY helps you, don’t forget to leave a star!

Made with 💙 by dvtlab

📧 Contact: dvtlab@outlook.com