AI-Powered Intelligent Segmentation Plugin

Transform your digital art workflow with cutting-edge AI segmentation technology. Smart Segments leverages the power of SAM2 (Segment Anything Model 2) to provide intuitive, one-click object detection and selection in Krita.

• 🧠 AI-Powered Segmentation - Advanced machine learning for precise object detection
• 🎯 Interactive Selection - Click-to-select any object in your artwork
• ⚡ Real-time Preview - Instant visual feedback as you work
• 🔧 Smart Tools Integration - Seamless integration with Krita's native tools
• 🎨 Artist-Friendly - Designed by artists, for artists
• 🚀 One-Click Setup - Automated installation and configuration

Built and tested on Ubuntu 24.04 LTS, with potential cross-platform compatibility for:

• 🐧 Linux (Ubuntu, Debian, Fedora, Arch)
• 🪟 Windows 10/11
• 🍎 macOS (experimental)

sudo apt install libgtk-4-dev libadwaita-1-dev git blueprint-compiler gettext libxml2-utils libgtk4-layer-shell-dev

Note: While developed on Ubuntu, the plugin architecture supports cross-platform deployment. Windows and macOS compatibility may require additional testing.

1. Open Krita
2. Navigate to Tools → Scripts → Import Python Plugin from Web
3. Enter this GitHub repository URL:

   https://github.com/a904guy/Smart_Segments/releases/download/1.0.2/SmartSegments_v1.0.2.zip
   

4. Click OK and wait for download to complete
5. Restart Krita
6. Go to Settings → Configure Krita → Python Plugin Manager
7. Enable "Smart Segments" plugin
8. Restart Krita again
9. Access via Tools → Smart Segments

1. Download the latest release ZIP from GitHub Releases
2. Open Krita
3. Navigate to Tools → Scripts → Import Python Plugin from Zip File
4. Select the downloaded ZIP file
5. Restart Krita
6. Go to Settings → Configure Krita → Python Plugin Manager
7. Enable "Smart Segments" plugin
8. Restart Krita again
9. Access via Tools → Smart Segments

./dev-install.sh

# Create Krita plugin directory
mkdir -p ~/.local/share/krita/pykrita

# Create symbolic links for development
ln -sf "$(pwd)/smart_segments" ~/.local/share/krita/pykrita/smart_segments
ln -sf "$(pwd)/smart_segments.desktop" ~/.local/share/krita/pykrita/smart_segments.desktop

See Smart Segments in action with these screenshots from the development process:

Development Mode: The plugin uses symbolic links, so code changes are immediately available after restarting Krita.

📦 Smart Segments
├── 🎨 smart_segments/             # Main plugin source
├── 🧪 tests/                     # Test suite
├── 📚 documents/                 # Documentation
├── 🎁 resources/                 # Icons and assets
├── 🔧 dev-*.sh                   # Development scripts
└── 📋 create_plugin_package.py   # Distribution builder

If you have the Snap version of Krita, you'll need to switch to the APT version:

# Remove snap version
sudo snap remove krita

# Install APT version with Python support
sudo apt install krita python3-pyqt5

# System check
./dev-utils.sh version

# Plugin verification
ls -la ~/.local/share/krita/pykrita/smart_segments

# Real-time log monitoring
tail -f ~/.local/share/krita/krita.log

When you first launch Smart Segments, the Setup Wizard automatically handles everything:

graph LR
    A[Launch Plugin] --> B[System Check]
    B --> C[Create Virtual Env]
    C --> D[Download AI Models]
    D --> E[Install Dependencies]
    E --> F[Test Segmentation]
    F --> G[Ready to Use! 🎉]

📡 Internet Required: First setup requires internet connection for downloading AI models (~1GB)

• PyTorch - Deep learning framework
• SAM2 Models - AI segmentation models
• NumPy - Numerical computing
• OpenCV - Computer vision utilities
• Pillow - Image processing

• Base Plugin: ~50MB
• AI Models: ~1GB
• Tooling/Dependencies: ~4-5GB
• Total: ~5-6GB

We welcome contributions! See our contributing guidelines for details.

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

• Krita Foundation - For the amazing open-source digital painting software
• Meta AI - For the incredible SAM2 model
• PyTorch Team - For the powerful deep learning framework