Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Djdefrag/QualityScaler
QualityScaler - image/video AI upscaler app
https://github.com/Djdefrag/QualityScaler
amd anime compression-artifact-reduction deep-learning directx-12 gui-application intel manga noise-reduction nvidia opencv python python3 pytorch pytorch-directml super-resolution video video-processing windows
Last synced: 11 days ago
JSON representation
QualityScaler - image/video AI upscaler app
- Host: GitHub
- URL: https://github.com/Djdefrag/QualityScaler
- Owner: Djdefrag
- License: mit
- Created: 2022-04-03T14:42:56.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-04-16T09:39:42.000Z (7 months ago)
- Last Synced: 2024-04-21T11:25:56.455Z (7 months ago)
- Topics: amd, anime, compression-artifact-reduction, deep-learning, directx-12, gui-application, intel, manga, noise-reduction, nvidia, opencv, python, python3, pytorch, pytorch-directml, super-resolution, video, video-processing, windows
- Language: Python
- Homepage:
- Size: 212 MB
- Stars: 1,718
- Watchers: 30
- Forks: 117
- Open Issues: 28
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
QualityScaler - image/video AI upscaler app
## What is QualityScaler?
Qualityscaler is a Windows app powered by AI to enhance, upscale and de-noise photographs and videos.
## How to use QualityScaler?
[movie_max_vp9.webm](https://github.com/user-attachments/assets/4a1984be-c8c8-464b-a3d1-43445635b165)
## Other AI projects.🤓
- https://github.com/Djdefrag/RealScaler / RealScaler - image/video AI upscaler (Real-ESRGAN)
- https://github.com/Djdefrag/FluidFrames.RIFE / FluidFrames.RIFE - video AI frame generation
## Credits.
- BSRGAN - https://github.com/cszn/BSRGAN
- Real-ESRGAN - https://github.com/xinntao/Real-ESRGAN
- IRCNN - https://github.com/lipengFu/IRCNN
## Citations. ❤
- https://80.lv/articles/80-level-digest-great-ai-powered-tools-for-upscaling-images/
- https://timesavervfx.com/ai-upscale/
## How is made. 🛠
QualityScaler is completely written in Python, from backend to frontend.
External packages are:
- AI
- Pytorch (https://github.com/pytorch/pytorch)
- onnx (https://github.com/onnx/onnx)
- onnxconverter-common (https://github.com/microsoft/onnxconverter-common)
- onnxruntime-directml (https://github.com/microsoft/onnxruntime)
- GUI
- customtkinter (https://github.com/TomSchimansky/CustomTkinter)
- Image/video
- OpenCV (https://github.com/opencv/opencv)
- moviepy (https://github.com/Zulko/moviepy)
- Packaging
- PyInstaller (https://github.com/pyinstaller/pyinstaller)
## Make it work by yourself. 👨💻
Prerequisites.
- Python installed on your pc (https://www.python.org/downloads/release/python-3119/)
- VSCode installed on your pc (https://code.visualstudio.com/)
- AI models downloaded (https://gofile.io/d/hIOmMc)
- FFMPEG.exe downloaded (https://www.gyan.dev/ffmpeg/builds/) RELEASE BUILD > ffmpeg-release-essentials.7z
Getting started.
- Download the project on your PC (Green button Code > Download ZIP)
- Extract the project from the .zip
- Extract the AI models files in /AI-onnx folder
- Extract FFMPEG.exe in /Assets folder
- Open the project with VSCode (Drag&Drop the project directory on VSCode)
- Click on QualityScaler.py from left bar (VSCode will ask to install Python plugins)
- Install dependencies. In VSCode there is the "Terminal" panel, click there and execute the command "pip install -r requirements"
- Close VSCode and re-open it (this will refresh all the dependecies installed)
- Click on the "Play button" in the upper right corner of VSCode
## Requirements. 🤓
- Windows 11 / Windows 10
- RAM >= 8Gb
- Any Directx12 compatible GPU with >= 4GB VRAM
## Features.
- [x] Elegant and easy to use GUI
- [x] Image and Video upscale
- [x] Multiple GPUs support
- [x] Compatible images - jpg, png, tif, bmp, webp, heic
- [x] Compatible video - mp4, wemb, mkv, flv, gif, avi, mov, mpg, qt, 3gp
- [x] Automatic image tilling to avoid gpu VRAM limitation
- [x] Resize image/video before upscaling
- [x] Interpolation beetween original file and upscaled file
- [x] Video upscaling STOP&RESUME
- [x] PRIVACY FOCUSED - no internet connection required / everything is on your PC
## Next steps. 🤫
- [x] 1.X versions
- [x] Switch to Pytorch-directml to support all Directx12 compatible gpu (AMD, Intel, Nvidia)
- [x] New GUI with Windows 11 style
- [x] Include audio for upscaled video
- [x] Optimizing video frame resize and extraction speed
- [x] Multi GPU support (for pc with double GPU, integrated + dedicated)
- [x] Python 3.10 (expecting ~10% more performance)
- [x] 2.X versions
- [x] New, completely redesigned graphical interface based on @customtkinter
- [x] Upscaling images and videos at once (currently it is possible to upscale images or single video)
- [x] Upscale multiple videos at once
- [x] Choose upscaled video extension
- [x] Interpolation between the original and upscaled image/video
- [x] More Interpolation levels (Low, Medium, High)
- [x] Show the remaining time to complete video upscaling
- [x] Support for SRVGGNetCompact AI architecture
- [x] Metadata extraction and application from original file to upscaled file (via exiftool)
- [ ] 3.X versions
- [x] New AI engine powered by onnxruntime-directml (https://github.com/microsoft/onnxruntime))
- [x] Python 3.11 (performance improvements)
- [x] Python 3.12 (performance improvements)
- [x] Display images/videos upscaled resolution in the GUI
- [x] Updated FFMPEG to version 7.x (latest release)
- [x] Saving user settings (AI model, GPU, CPU etc.)
- [x] Video multi-threading AI upscale
- [x] Video upscaling STOP&RESUME
## Some Example.
#### Videos
https://user-images.githubusercontent.com/32263112/209139639-2b123b83-ac6e-4681-b94a-954ed0aea78c.mp4
#### Images
