Somewhere We Live in Little Loops

A collaboration with a machine-learning based AI technology of next-frame prediction. The computerised synthesis of images by a neural network learns video clips and then guesses how they will continue, recreating them in the process. The reborn video images are then woven together into an audio-visual poem, revealing both computer and human cognition processes and raising reflections regarding the implications and future of this new technology.

Metadata from the source database.