Idiom is a living, breathing sculptural mask developed by Pierre Huyghe, blending technology, biology, and performance into a haunting presence. Worn by a performer and shaped like a neural web, the mask interprets its surroundings using embedded machine vision.
As a developer on the project, I worked closely with the artist’s studio to research and implement the computer vision system at the heart of the work. Using a Raspberry Pi and the YOLO (You Only Look Once) object detection framework, we optimized the system for real-time inference directly on-device. The result: a fully self-contained, wearable vision system capable of identifying and reacting to environmental stimuli without relying on cloud processing or external power.
My contributions focused on lightweight architecture, performance tuning, and system stability—ensuring the piece could perform in diverse gallery settings, untethered.
