BlockFusion: Generating Expandable and High-Quality 3D Scenes using Latent Tri-plane Extrapolation

BlockFusion is a diffusion-based model that generates 3D scenes as unit blocks and seamlessly incorporates new blocks to extend the scene. It leverages a latent tri-plane representation and a denoising diffusion process to produce diverse, geometrically consistent, and unbounded large 3D scenes with high-quality shapes.