VisionMIDI Live

VisionMIDI Live transforms any video feed into an expressive musical instrument using cutting-edge AI object detection. Point a camera at busy traffic intersections, crowded theatres, birds circling in the sky, or any dynamic scene, and watch as detected objects trigger notes on a configurable virtual piano overlay, creating music in real-time. With YOLOv11 object detection and intelligent tracking, the natural world becomes your orchestra.

VisionMIDI Live opens up limitless possibilities for sound installations, generative music systems, and performances that celebrate the inherent musicality found in the movement patterns of our world - from the structured chaos of urban environments to the natural rhythms of the living world.

Vision MIDI Live working on a London street scene.