VUG - Video Upscaling GUI

VUG is a cross-platform GUI application for video upscaling using Real-ESRGAN neural networks. Built with Python and PySide6, managed with BeeWare's Briefcase.

Features

• User-friendly graphical interface
• Multiple Real-ESRGAN model support
• Adjustable scaling factors (2x, 3x, 4x)
• Configurable batch processing
• Real-time progress logging
• Task duration tracking
• Preserves original audio tracks
• Supports common video formats (MP4, MOV, AVI, MKV, etc.)

Requirements

Core Dependencies:

• Python 3.8+
• FFmpeg (system PATH)
• realesrgan-ncnn-vulkan (system PATH)

Installation

Binary:

You can find what binary builds I am able to build in the releases.

From Source:

git clone https://codeberg.org/Melon-Bread/VUG
cd VUG
pip install briefcase # Can also use pipx

Usage

1. Ensure required binaries are in your PATH:

   • ffmpeg
   • realesrgan-ncnn-vulkan

2. Launch VUG:

   briefcase run

3. GUI Interface:

   • Select input video/directory
   • Choose output directory
   • Configure upscaling parameters
   • Monitor progress via real-time logs

Supported Models

• realesr-animevideov3
• realesrgan-x4plus
• realesrgan-x4plus-anime
• realesrnet-x4plus

Contributing

1. Fork the repository
2. Create feature branch
3. Submit merge request
4. Ensure code complies with GPLv3 license

Third-Party Tools

• FFmpeg: LGPL-2.1-or-later
• Real-ESRGAN-ncnn-vulkan: MIT License
• PySide6: LGPL-3.0-only
• Briefcase: BSD-3-Clause