Xoromancy

Two Carnegie Mellon students built the first real-time tool that let the human body steer a neural network's imagination. Xoromancy mapped a visitor's hand movements, tracked by a Leap Motion sensor, directly into the latent space of BigGAN, DeepMind's image generator. Gray Crawford announced it in February 2019: "like divination via movement."

The insight was that BigGAN's thousands of input variables were hopeless to navigate with sliders, but the human body is itself a high-dimensional control system. Tiwari built the pipeline in Unity and TensorFlow. Years before anyone typed a prompt, Xoromancy posed the question the field would inherit: not whether a model can generate an image, but how a human directs it.