Sparse auto-regressive modeling for scene generation from multi-view images
This work introduces SPAR3S, a sparse voxel-aligned 3D latent generative model for conditional scene completion without requiring ground-truth 3D data for supervision, and trains a masked autoregressive transformer that jointly models voxel occupancy and latent token values, enabling efficient and spatially consistent...
Thomas Lucas, Maxime Pietrantoni, Philippe Weinzaepfel et al.
· 1 citation