Bolt3D: Generating 3D Scenes in Under 7 Seconds

Bolt3D, a collaborative effort from Google Research, VGG, and Google DeepMind, generates realistic 3D scenes in a mere 6.25 seconds on a single GPU. The model uses a multi-view diffusion model to generate scene appearance and geometry, then regresses splatter images using a Gaussian head. Finally, it combines 3D Gaussians from multiple splatter images to form the complete 3D scene. Supporting one or more input images, Bolt3D generates unobserved scene regions without reprojection or inpainting, showcasing a significant leap in 3D scene generation speed.