Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/nenotriple/pixel8
Pixelate an image using various methods and conditions.
https://github.com/nenotriple/pixel8
edit-image editing-image image-manipulation image-processing pixel-art stable-diffusion
Last synced: about 2 months ago
JSON representation
Pixelate an image using various methods and conditions.
- Host: GitHub
- URL: https://github.com/nenotriple/pixel8
- Owner: Nenotriple
- Created: 2024-09-16T09:16:07.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2024-09-17T02:32:15.000Z (5 months ago)
- Last Synced: 2024-09-17T11:20:25.226Z (5 months ago)
- Topics: edit-image, editing-image, image-manipulation, image-processing, pixel-art, stable-diffusion
- Language: Python
- Homepage:
- Size: 165 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Pixel8
Pixelate an image using various methods and conditions.
# 📝 Usage
> [!NOTE]
> - Supported image types: `.png` `.webp` `.jpg` `.jpeg` `.jpg_large` `.bmp` `.jfif` `.tiff` `.tif` `.ico`
This tool is designed to be used with AI-generated images in the pixel art style. It's common for these images to be generated with *mushed* colors and/or sloppy pixel shaping. Processing those images with this tool can enhance the pixel art style, and improve image quality.
---
# 💡 Tips and Features
Add to custom palette list...
1) Create a folder named `Palettes` in the same directory as the app.
2) Place any `PNG` image into the `Palettes` folder.
3) Restart the app and it will scan the `Palettes` folder for new images.
4) Palettes will appear using their filename in the custom palette drop-down menu.
Drag-and-Drop files into input fields...
Use another image as the color palette...
Pick from a list of pre-defined color palettes...
Normal and Blend color modes...
# 🚩 Requirements
> [!TIP]
> You don't need to worry about any requirements with the Windows [💾portable/executable](https://github.com/Nenotriple/Pixel8/releases?q=executable&expanded=true) version.
Python requirements...
**Python 3.10+**
You will need `Pillow`, `NumPy`, `TkinterDnD2`, `scikit-learn`, and `matplotlib`:
- `pip install pillow numpy tkinterdnd2 scikit-learn matplotlib`
Or use the included `requirements.txt` when creating your virtual enviroment.
# 📜 Version History
[💾v1.01](https://github.com/Nenotriple/Pixel8/releases/tag/v1.01)
Click here to view release notes for v1.01
- New:
- `Batch Mode`: Allows input/output entries to accept folder paths.
- Processes all supported images in subfolders while maintaining folder structure.
- Before and After preview images are now displayed. *(When not in Batch mode)*
- The Before image also works as a drop target to set the input image path.
- Right-click on the palette in the UI to quickly save the palette image with `Save As...`.
- Additional palettes can be used by placing `.PNG` files in a `Palettes` folder in the app's parent directory.
- Supports up to 256 colors per palette.
- `Sharpen` Setting: Enhances image sharpness to improve pixel contrast and color.
- Sharpen value range: From -100, to 100.
- You can now choose between two color transfer modes: `Normal` and `Blend`.
- `Normal` mode assigns each pixel to the nearest color in the palette. *(Previous behavior)*
- `Blend` mode blends the two nearest colors based on their distance.
- Added `Open` buttons to each input entry to quickly open the selected directory.
- Fixed:
- Fixed error when processing non-RGB format images. Alpha channels are now preserved.
- Fixed issue where color palettes included colors not present in the input image.
- Unsupported file types can no longer be drag-and-dropped into input fields.
- Other changes:
- Image processing is ~70% faster (Kmeans/Normal) due to improved pixel-color handling.
- Setting `Image Downscale` to `0`, or `1`, now disables downscaling.
- Palette images are now saved as 1x1 grids, expanding in width.
- Palettes contain only unique colors, sorted by brightness.
- Unfortunately, it's not easy to prevent very *similar* colors from being included.
- Maximum number of colors changed from 999 to 256.
- Supported image file types now include: `.jfif`, `.tiff`, `.tif`, and `.ico`.
- All spinboxes can be adjusted with the mouse-wheel.
- Tooltips now mostly appear when hovering the mouse over label text and not input widgets, reducing their annoyance.
- Numerous small tweaks, fixes, and improvements in both logic and UI.
- The app has been internally restructured and organized.