Diffusion$^2$: Dynamic 3D Content Generation via Score Composition of Orthogonal Diffusion Models

arXiv:2404.02148v2 Announce Type: replace
Abstract: Recent advancements in 3D generation are predominantly propelled by improvements in 3D-aware image diffusion models which are pretrained on Internet-scale image data and fine-tuned on massive 3D data, offering the capability of producing highly consistent multi-view images. However, due to the scarcity of synchronized multi-view video data, it is impractical to adapt this paradigm to 4D generation directly. Despite that, the available video and 3D data are adequate for training video and multi-view diffusion models …

abstract arxiv capability consistent content generation cs.cv data diffusion diffusion models dynamic however image image data image diffusion images improvements internet massive scale type via view