Agentic Video Editor

An AI-powered video editing application with a Qt6/C++ frontend and Python backend, featuring node-based workflows and agent-driven automation.

Features

Core Video Editing

• Timeline Editor: Multi-track timeline with drag-and-drop support
• Video Player: Hardware-accelerated playback with frame-accurate seeking
• Clip Management: Split, trim, move, and arrange clips with ease
• Undo/Redo System: Full command pattern implementation for all operations

Node-Based Workflows

• Visual Node Graph: Create complex video processing pipelines
• AI-Powered Nodes: Text-to-video, image-to-video, style transfer, and more
• Custom Nodes: Extensible architecture for adding new processing nodes
• Real-time Preview: See changes as you build your workflow

AI Agent Console

• Natural Language Commands: Edit videos using plain English instructions
• Smart Suggestions: AI-powered recommendations for edits and effects
• Workflow Automation: Let the agent build node graphs automatically
• Speech-to-Text: Integrated Whisper model for transcription

Advanced Features

• Proxy Media: Automatic proxy generation for smooth 4K+ editing
• Preview Cache: Intelligent caching for real-time playback
• Audio Waveform: Visual audio editing with waveform display
• Performance Profiler: Built-in profiling for optimization

Tech Stack

Frontend

• Qt6 - Cross-platform GUI framework
• C++17 - High-performance core
• QMediaPlayer - Video playback
• QGraphicsView - Node graph and timeline rendering

Backend

• Python 3.8+ - AI and processing backend
• FastAPI - REST API server
• Whisper - Speech recognition
• Diffusers - AI model inference

Build System

• CMake 3.16+ - Build configuration
• Ninja - Fast build tool

Project Structure

.
├── src/
│   ├── ui/              # Qt/C++ frontend components
│   │   ├── mainwindow.cpp/h
│   │   ├── timeline_widget.cpp/h
│   │   ├── video_player.cpp/h
│   │   ├── node_widget.cpp/h
│   │   └── agent_console.cpp/h
│   ├── engine/          # Core engine components
│   │   ├── timeline_model.cpp/h
│   │   ├── video_decoder.cpp/h
│   │   ├── render_pipeline.cpp/h
│   │   └── undo_manager.cpp/h
│   └── ai_nodes/        # Python backend
│       ├── node_server.py
│       ├── requirements.txt
│       └── nodes/       # AI processing nodes
├── workflows/           # Pre-built workflow templates
├── build/              # Build output (gitignored)
└── CMakeLists.txt      # CMake configuration

Building from Source

Prerequisites

• Qt6 (Core, Widgets, Multimedia)
• CMake 3.16+
• Python 3.8+
• C++17 compatible compiler

Windows

# Clone the repository
git clone https://github.com/Hemanthc-dotcom/Kdenagent.git
cd Kdenagent

# Create build directory
mkdir build
cd build

# Configure with CMake
cmake .. -G Ninja

# Build
ninja

# Run
.\AgenticEditor.exe

Linux/macOS

# Clone the repository
git clone https://github.com/Hemanthc-dotcom/Kdenagent.git
cd Kdenagent

# Create build directory
mkdir build && cd build

# Configure with CMake
cmake .. -G Ninja

# Build
ninja

# Run
./AgenticEditor

Python Backend Setup

# Create virtual environment
python -m venv .venv

# Activate (Windows)
.venv\Scripts\activate

# Activate (Linux/macOS)
source .venv/bin/activate

# Install dependencies
pip install -r src/ai_nodes/requirements.txt

# The backend starts automatically when you run the application

Usage

Basic Editing

1. Import Media: Drag and drop videos or use File > Import
2. Add to Timeline: Drag clips from the media pool to the timeline
3. Edit: Use the toolbar tools to cut, trim, and arrange clips
4. Preview: Play back your edit in the video player
5. Export: Render your project to various formats

Node Workflows

1. Open the Node Editor from the View menu
2. Add nodes from the palette (right-click in the canvas)
3. Connect nodes by dragging between ports
4. Configure node parameters in the properties panel
5. Execute the workflow to process your media

Agent Console

1. Open the Agent Console from the View menu or press Ctrl+Shift+A
2. Type natural language commands like:
   • "Add a fade-in effect to the first clip"
   • "Create a text overlay saying 'Hello World'"
   • "Generate B-roll for this section"
3. The AI will interpret your command and execute the appropriate actions

Keyboard Shortcuts

Shortcut	Action
Space	Play/Pause
J	Rewind
K	Pause
L	Forward
I	Set In Point
O	Set Out Point
Ctrl+Z	Undo
Ctrl+Shift+Z	Redo
Ctrl+S	Save Project
Delete	Delete Selected
Ctrl+Shift+A	Open Agent Console

Available AI Nodes

• Text2Video: Generate video from text descriptions
• Image2Video: Animate static images
• Video2Video: Transform video style using AI
• StyleTransfer: Apply artistic styles to footage
• SuperResolution: Upscale video resolution
• LipSync: Synchronize lips to audio
• AudioReactive: Create visuals that react to audio
• SceneDetection: Automatically detect scene changes
• HighlightDetection: Find exciting moments in footage
• CaptionGenerator: Auto-generate subtitles
• BrollGenerator: Generate contextual B-roll footage
• MagicEraser: Remove unwanted objects from video

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

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

Acknowledgments

• Qt Framework for the excellent cross-platform GUI toolkit
• OpenAI Whisper for speech recognition
• Hugging Face Diffusers for AI model inference
• The open-source community for various libraries and tools

Support

For issues, questions, or suggestions, please open an issue on GitHub.

---

Made with ❤️ by Hemanth