Open Source · Python 3.10+ · GUI + CLI

Convert image sequences
into ASCII art frames

A pipeline-ready ASCII renderer for premultiplied RGBA PNG sequences. Export white-on-black, black-on-white, or full-colour — with custom branded text woven through the artwork.

Download View on GitHub
ascii_preview — live
3
render styles
∞
frames per batch
2
pip packages
GUI
+ CLI included
What it does

Built for motion graphics pipelines

Processes entire PNG sequences, not just single images. Drop a folder of frames, get back a folder of ASCII renders.

🎞️

Batch frame processing

Drop an entire folder of PNGs and receive a matching folder of ASCII renders — numbered identically, ready to re-import.

🎨

Three render styles

White-on-black, black-on-white, and full-colour — all three generated in a single pass from one source scan.

🔤

Custom branded text

Scatter any string — a brand name, logo, year — intact through the subject. Density is fully adjustable.

📐

Exact output dimensions

Output PNGs are always resized to exactly match the source frame. No pixel-dimension surprises or aspect ratio drift.

🔲

Premultiplied alpha

Correctly unpremultiplies RGBA before converting — subject edges come out clean, not dark-fringed or blurred.

🖥️

GUI + CLI

Run headlessly in a terminal or launch the tkinter GUI for a live settings panel with a real-time progress bar and log.

CLI demo

See it run

Point it at any folder of PNG frames and it handles the rest.

Terminal — bash
GUI

No terminal required

The tkinter interface exposes every setting — browse for your folder, tweak, and hit Convert.

ASCII Art Converter · Polka Dot Post
Input Folder
/frames/shot_001/
Browse
Cell & Font
Cell W (px)
6
Cell H (px)
12
Font size (pt)
12
Custom Strings
BRAND, PDP, ASCII, 2026
Density
0.50
Output Style
All
Mono
W/B
B/W
Colour
Alpha & Flags
Alpha threshold
30
✓
Premultiplied alpha
▶ Convert
■ Cancel
Output Log
────────────────────────────────── Folder : /frames/shot_001/ Frames : 240 Cell : 6×12px → font 12pt Style : all ────────────────────────────────── Font: consola.ttf Char px: 7w × 14h ────────────────────────────────── [ 1/240] frame_001 ✓ [ 2/240] frame_002 ✓ [ 3/240] frame_003 ✓ ... [173/240] frame_173 ✓
Getting started

Up and running in minutes

Requires Python 3.10+ and two pip packages — nothing else.

1

Clone or download

git clone https://github.com/harv12321/ascii-art-converter.git

2

Install dependencies

pip install Pillow numpy — that's it, no other dependencies.

3

Run the CLI

python ascii_converter_2.py /path/to/png/folder — results appear in ascii_output/ alongside your frames.

4

Or launch the GUI

python ascii_converter_gui.py — browse for your folder, tweak any setting, and hit Convert.

Configuration

All settings at a glance

Edit the CONFIG block at the top of ascii_converter_2.py, or use the GUI for live adjustments.

Setting Default Description
CELL_W / CELL_H 6 / 12 Source pixels per ASCII character. Keep H ≈ 2×W to match monospace aspect ratios. Smaller = more detail.
FONT_SIZE 12 Point size of the rendered monospace font in the output PNG.
FONT_PATH None (auto) Explicit path to a .ttf / .otf file. Auto-detects Consolas, Courier, DejaVu, Menlo across platforms.
CUSTOM_STRINGS ["BRAND", "PDP", "ASCII", "2026"] Strings scattered intact through the subject — each is always placed as a consecutive unit, never split.
CUSTOM_DENSITY 0.5 Probability a subject character is drawn from your strings vs the ASCII ramp. 0 = never, 1 = always.
STYLE "all" "white_on_black", "black_on_white", "colour", "both", or "all".
ALPHA_THRESHOLD 30 Cells with mean alpha below this value are treated as background (range 0–255).
IS_PREMULTIPLIED True Set False for subject-on-black PNGs without a real alpha channel (e.g. black-background renders).
OUTPUT_FOLDER "ascii_output" Name of the subfolder created inside your input folder. Style subfolders are created within it.