ASCILINE — turning video into ASCII as a render style
Can I turn video into ASCII art as a deliberate visual style — with control over the character set, gain, gamma, and color-vs-mono — using my own code instead of a third-party tool with murky licensing?
ASCII output is paradoxical in weight: as plain text it's tiny (~1 MB), but rendered back into pixels as a video or GIF it balloons (~200 MB). That single fact changes how you'd ever publish it. Writing my own exporter kept the licensing clean — no third-party tool's terms attached to my output — and gave me the dials (charset, gain, gamma, mono) to actually shape the look.
It's a working exporter and an aesthetic experiment. Parked as a style tool; the outputs travel best as GIFs rather than raw video, so that's how they'd show up in my work.
What it is
ASCILINE is my own script (export_ascii_video.py) that converts video into ASCII art —
the image rebuilt out of text characters. It's not a novelty filter; it's a controllable
render style, with knobs for the character set, gain, gamma, and color versus monochrome.
The interesting wrinkle
ASCII has a split personality when it comes to size:
- As text, a whole clip is featherweight — around 1 MB.
- As pixels (a rendered video or GIF of that same ASCII), it's heavy — around 200 MB.
Same content, two-hundred-fold difference. That decides everything about how you publish it, which is why the outputs live as GIFs rather than full video.
Why I built it instead of using a tool
Two reasons. First, licensing: using my own code means no third-party tool's terms ride along on whatever I make — important if any of it ever supports marketing. Second, control: the gain/gamma/charset settings let me tune the texture instead of accepting a preset.
Where it stands
It works. It's a style in my kit now — a way to "render" footage as text. There's a conceptual thread I like here too: ASCII as another way to render a prototype, not just decorate it.