RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generation
This addresses the need for 3D understanding in AI, offering a novel method for 3D generation and inference with only 2D supervision, which is incremental but impactful for computer vision and graphics applications.
The paper tackles the problem of enabling image diffusion models to perform 3D tasks like view-consistent generation and single-view reconstruction, achieving competitive performance on datasets such as FFHQ, AFHQ, ShapeNet, and CLEVR.
Diffusion models currently achieve state-of-the-art performance for both conditional and unconditional image generation. However, so far, image diffusion models do not support tasks required for 3D understanding, such as view-consistent 3D generation or single-view object reconstruction. In this paper, we present RenderDiffusion, the first diffusion model for 3D generation and inference, trained using only monocular 2D supervision. Central to our method is a novel image denoising architecture that generates and renders an intermediate three-dimensional representation of a scene in each denoising step. This enforces a strong inductive structure within the diffusion process, providing a 3D consistent representation while only requiring 2D supervision. The resulting 3D representation can be rendered from any view. We evaluate RenderDiffusion on FFHQ, AFHQ, ShapeNet and CLEVR datasets, showing competitive performance for generation of 3D scenes and inference of 3D scenes from 2D images. Additionally, our diffusion-based approach allows us to use 2D inpainting to edit 3D scenes.