Proper Pixel Art

by Kenneth Allen

Noisy, high resolution

→

Clean, true-resolution pixel art

Summary

Converts noisy, high-resolution pixel-art-style images (from generative models or low-quality web uploads) into clean, true-resolution assets. Such images often have a non-uniform grid and random artifacts, so standard downsampling fails — the usual alternatives are naive downscaling or redrawing the asset pixel by pixel. This tool automates the recovery instead. Videos and GIFs are supported too via ppa-video.

Contents

• Installation
  • Install From PyPI
  • Install from source
• Usage
  • Web Interface
  • CLI
  • Videos and GIFs
  • Use Without Cloning
  • Python API
  • Configuration file
• Examples
• Real Images To Pixel Art
• Algorithm

Installation

Install From PyPI

pip install proper-pixel-art  # CLI and Python API
pip install "proper-pixel-art[web]"  #  Include the local web UI

Or with uv:

uv add proper-pixel-art  # CLI and Python API
uv add proper-pixel-art --extra web  # Include the local web UI

Install from source

git clone [email protected]:KennethJAllen/proper-pixel-art.git
cd proper-pixel-art
uv sync --extra web

Usage

First, obtain a source pixel-art-style image (e.g. from a generative model such as OpenAI's gpt-image-2, or a web upload of pixel art).

The examples below assume you installed via pip install or uv add (commands are on your PATH). If you installed from source with uv sync, prefix each command with uv run (e.g. uv run ppa ...).

Web Interface

Try it live in your browser, no install required, on Hugging Face Spaces.

To run the same interface locally:

ppa-web
# Opens http://127.0.0.1:7860

CLI

ppa <input_path> -o <output_path> -c <num_colors> -s <result_scale> [-t]

Options

Example

ppa assets/blob/blob.png -c 16 -s 25

Note: --colors is the parameter most likely to need tuning. See the option table above.

Videos and GIFs

Pixelate animations (e.g. from video models such as Sora) with ppa-video. The pixel mesh and color palette are computed once from sampled frames and applied to every frame, so the animation stays consistent with no flicker.

ppa-video <input.mp4|input.gif> -o <output_path> -c <num_colors> [-f mp4|gif]

It accepts the same pixelation options as ppa (see the table above), plus:

GIF input is decoded with full frame compositing (variable-size delta frames, per-frame durations, and transparency are preserved). GIF output uses a single global palette.

From Python:

from proper_pixel_art.video import pixelate_video

pixelate_video('input.mp4', 'output.gif', num_colors=16)

Use Without Cloning

Web Interface (without cloning)

uvx --from "proper-pixel-art[web]" ppa-web

CLI (without cloning)

uvx --from "proper-pixel-art" ppa <input_path>

Python API

For Python developers who want to integrate this tool into their own code.

from PIL import Image
from proper_pixel_art import pixelate

image = Image.open('path/to/input.png')
result = pixelate(image, num_colors=16)
result.save('path/to/output.png')

Parameters

These mirror the CLI options above.

• image : PIL.Image.Image — the image to pixelate.
• num_colors : int — colors in result (1-256), or 0 to skip quantization. Most likely to need tuning.
• initial_upscale_factor : int — upscale the input first; may help detect lines.
• scale_result : int — upscale the result; 1 = no scaling.
• transparent_background : bool — if True, flood-fill each corner with transparent alpha.
• intermediate_dir : Path | None — save visualizations of intermediate steps (for development).
• pixel_width : int — pixel width in the input, or 0 to detect automatically.
• config : PixelateConfig | None — a bundle of every tunable parameter, including the deeper mesh-detection (Canny, Hough, line clustering) and color (alpha/transparency thresholds, quantization method, color binning) settings not exposed as direct arguments. Load one with PixelateConfig.from_yaml(path). Explicit arguments override matching values in config.

Returns

A PIL image with true pixel resolution and quantized colors.

Configuration file

All tunable parameters can be collected in a YAML file so you can fine-tune the algorithm without changing code. See config.example.yaml for the full list of keys with their defaults. Any key you omit falls back to the default, so partial files are fine.

from PIL import Image
from proper_pixel_art import pixelate
from proper_pixel_art.config import PixelateConfig

config = PixelateConfig.from_yaml('config.yaml')
result = pixelate(Image.open('input.png'), config=config)

From the CLI, pass --config. Flags given explicitly override values from the file:

ppa input.png --config config.yaml      # use the file
ppa input.png --config config.yaml -c 8 # but override num_colors to 8

Examples

The algorithm is robust. It performs well for images that are already approximately aligned to a grid.

Here are a few examples. A mesh is computed, where each cell corresponds to one pixel.

Bat

• Generated by GPT-4o.

Noisy, High Resolution

Mesh

True Pixel Resolution

Ash

• Screenshot from Google images of Pokemon asset.

Noisy, High Resolution

Mesh

True Pixel Resolution

Demon

• Original image generated by GPT-4o.

Noisy, High Resolution

Mesh

True Pixel Resolution

Pumpkin

• Screenshot from Google Images of Stardew Valley asset. This is an adversarial example as the source image is both low quality and the object is round.

Noisy, High Resolution

Mesh

True Pixel Resolution

Real Images To Pixel Art

• This tool can also be used to convert real images to pixel art by first requesting a pixelated version of the original image from GPT-4o, then using the tool to get the true pixel-resolution image.

• Consider this image of a mountain

• Here are the results of first requesting a pixelated version of the mountain, then using the tool to get a true resolution pixel art version.

Noisy, High Resolution

Mesh

True Pixel Resolution

Algorithm

Here's a step-by-step overview, applied to this GPT-4o-generated blob:

• Note that this image is high resolution and noisy.

1. Trim the edges of the image and zero out pixels with more than 50% alpha.

   • This is to work around some issues with models such as GPT-4o not giving a perfectly transparent background.

2. Upscale by a factor of 2 using nearest neighbor.

   • This can help identify the correct pixel mesh.

3. Find edges of the pixel art using Canny edge detection.

4. Close small gaps in edges with a morphological closing.

5. Take the probabilistic Hough transform to get the coordinates of lines in the detected edges. Only keep lines that are close to vertical or horizontal giving some grid coordinates. Cluster lines that are closeby together.

6. Find the grid spacing by filtering outliers and taking the median of the spacings, then complete the mesh.

7. Quantize the original image to a small number of colors (see the num_colors tuning note above).

8. In each cell specified by the mesh, choose the most common color in the cell as the color for the pixel. Recreate the original image with one pixel per cell.

   • Result upscaled by a factor of $20 \times$ using nearest neighbor.