bubamara ocr by adriana rami, 2018, 100 editions

A desktop application with achromatopsic computer vision for interpreting images as texts, with ladybugs as alphabetic intermediary

"Curious what would happen if I could ask a computer to interpret an image as a crude sequence of letters, I wrote a program, BUBAMARA-OCR, that could infer a text based on Latin alphabet from an image using a computer vision algorithm. Ladybirds, chosen for their roles as predictive and fateful messengers, both biologically observed and reimagined by generative adversarial networks, were the intermediary between photo and alphabet  each letter having its own corresponding ladybird. Whatever ladybird  each section of the image looked like most determined the letter that the program would output."

[source quote](https://v1.left.gallery/downloads/bubamara-badlands-adriana-ramic.pdf)

offline, interactive, silent, macos and windows