Vitrus / Demos

World Descriptor

See how a compact model understands the world through your camera.

Quantized model · 418 MB

Camera frames are used only for this inference session and are not retained or used for training.

Visual detail