🎬 DVC Vapourize

Professional DV to HD upscaling tool using Vapoursynth with QTGMC deinterlacing and chroma enhance

Transform your legacy DV footage into pristine 4K ProRes with professional-grade deinterlacing, chroma cleanup, and neural network upscaling.

✨ Features

• 🚀 QTGMC Deinterlacing - Industry-standard motion-compensated deinterlacing
• 🧠 NNEDI3 Neural Upscaling - AI-powered 2× upscaling to HD (1920x1080)
• 🎯 Advanced Chroma Cleanup - Multi-pass temporal and spatial chroma artifact removal
• 🎨 Color Space Conversion - Professional color space handling for cinema workflows
• 📁 Batch Processing - Process entire directories with progress monitoring
• 🖥️ GUI & CLI - Both graphical interface and command-line operation
• ⚡ Optimized Pipeline - Multi-threaded processing with performance profiling
• 📊 Rich Progress Tracking - Real-time frame-by-frame progress with ETA

🎯 Designed For

• DV Footage (720×576 PAL, 720×480 NTSC)
• Legacy Video Formats (AVI, MKV, MP4, MXF, DV, HDV)
• Professional Post-Production workflows
• Archive Restoration projects

🚀 Quick Start

Prerequisites

• Python 3.8+
• FFmpeg (in system PATH)
• VapourSynth R57+

1. Environment Setup

# Create virtual environment
python -m venv dvc_vapourize_env

# Activate environment
# Windows:
dvc_vapourize_env\Scripts\activate
# macOS/Linux:
source dvc_vapourize_env/bin/activate

# Install Python dependencies
pip install -r requirements.txt

2. VapourSynth Setup

Install VapourSynth and required plugins:

# Install vsrepo (VapourSynth plugin manager)
pip install vsrepo

# Install required plugins
vsrepo install qtgmc
vsrepo install nnedi3
vsrepo install havsfunc
vsrepo install ffms2
vsrepo install lsmas
vsrepo install mvtools
vsrepo install vsdehalo

# Update all plugins
vsrepo upgrade-all

3. Run the Application

GUI Mode

python main_window.py

Command Line Mode

# Single file processing
python vs_pipeline.py --input "input_video.avi" --output "output_directory"

# Batch processing with test mode
python vs_pipeline.py --input "input_directory" --output "output_directory" --batch_mode --test_mode --test_frames 500

# Full batch processing with recursive subdirectory scanning
python vs_pipeline.py --input "input_directory" --output "output_directory" --batch_mode --recursive

📋 Command Line Options

Option	Description	Default
--input	Input file or directory path	Required
--output	Output directory path	Required
--batch_mode	Enable batch processing	Single file
--recursive	Process subdirectories	Disabled
--test_mode	Process limited frames for testing	Disabled
--test_frames	Number of frames in test mode	200

🔧 Processing Pipeline

Input DV Video → Rewrap to ProRes → VapourSynth Processing → Final ProRes 422 HQ
                        ↓                       ↓                        ↓
                Format Detection    ┌─ Deinterlacing (QTGMC)    DCI-P3 Color Space
                Color Space Setup   ├─ Chroma Cleanup           Professional Codec
                Audio Preservation  ├─ Square Pixel Conversion  4K Resolution
                                   ├─ NNEDI3 2× Upscaling      
                                   └─ Format Optimization       

Processing Stages

1. Analysis & Rewrapping - Detect format, rewrap to ProRes container
2. Deinterlacing - QTGMC with Placebo preset for maximum quality
3. Chroma Cleanup - Multi-pass temporal and spatial artifact removal
4. Pixel Aspect Correction - Convert DV to square pixels (1920×1080)
5. Neural Upscaling - NNEDI3 2× upscaling to 4K (3840×2160)
6. Final Encoding - ProRes 422 HQ with DCI-P3 color space

🎛️ Advanced Configuration

VapourSynth Scripts

• upscale.vpy - Main processing pipeline
• rewrap.py - Format conversion and analysis

Key Parameters

# QTGMC Settings
Preset="Placebo"          # Maximum quality
SourceMatch=3             # Advanced source matching
Lossless=2               # High quality mode
MatchEnhance=0.95        # Motion matching enhancement

# NNEDI3 Settings
nsize=6                  # 32×6 neighborhood
nns=4                    # 256 neurons
qual=2                   # Highest quality

📁 Supported Formats

Input Formats

• Containers: AVI, MKV, MP4, MOV, MXF, TS, M2TS, DV, HDV
• Codecs: DV, HDV, H.264, MPEG-2, ProRes
• Resolutions: 720×576 (PAL DV), 720×480 (NTSC DV), 1440×1080 (HDV)

Output Format

• Container: QuickTime MOV
• Codec: ProRes 422 HQ (10-bit 4:2:2)
• Resolution: 3840×2160 (4K UHD)
• Color Space: DCI-P3
• Audio: PCM 24-bit uncompressed

🛠️ Troubleshooting

Common Issues

VapourSynth Import Errors

# Ensure VapourSynth is properly installed
python -c "import vapoursynth; print('VapourSynth OK')"

# Check plugin installation
vsrepo installed

FFmpeg Not Found

# Verify FFmpeg installation
ffmpeg -version

# Add FFmpeg to system PATH

Memory Issues

• Reduce thread count in settings
• Enable test mode for large files
• Process smaller batches

Performance Optimization

• CPU: Use thread count = CPU cores
• RAM: 16GB+ recommended for 4K processing
• Storage: SSD recommended for temp files
• GPU: NVIDIA GPU optional for monitoring

📊 Performance Metrics

Typical processing times (DV → 4K):

• Test Mode (200 frames): ~2-5 minutes
• Full Video (25fps PAL): ~0.1-0.5× real-time
• Batch Processing: Automatic queue management

🤝 Contributing

1. Fork the repository
2. Create feature branch (git checkout -b feature/enhancement)
3. Commit changes (git commit -am 'Add enhancement')
4. Push to branch (git push origin feature/enhancement)
5. Create Pull Request

📄 License

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

🙏 Acknowledgments

• VapourSynth Team - Powerful video processing framework
• QTGMC by Vit - Industry-standard deinterlacing
• NNEDI3 - Neural network edge-directed interpolation
• FFmpeg - Universal media processing

📞 Support

• 📖 Documentation: Check the code comments and docstrings
• 🐛 Issues: Report bugs via GitHub Issues
• 💬 Discussions: Join GitHub Discussions for help

---

Developed and tested on Windows 11 | Compatible with macOS and Linux