Denoising Diffusion via Image-Based Rendering

TL;DR

This paper introduces a novel diffusion-based method for detailed 3D scene reconstruction and generation from 2D images, utilizing a new neural scene representation and a unified framework that outperforms existing approaches.

Contribution

The work presents IB-planes for efficient large-scale 3D scene representation, a diffusion model trained on 2D images without extra supervision, and a strategy to prevent trivial solutions in 3D reconstruction.

Findings

Superior results in 3D generation and view synthesis

Effective reconstruction of real-world and synthetic scenes

Unification of 3D reconstruction and generation in a single model

Abstract

Generating 3D scenes is a challenging open problem, which requires synthesizing plausible content that is fully consistent in 3D space. While recent methods such as neural radiance fields excel at view synthesis and 3D reconstruction, they cannot synthesize plausible details in unobserved regions since they lack a generative capability. Conversely, existing generative methods are typically not capable of reconstructing detailed, large-scale scenes in the wild, as they use limited-capacity 3D scene representations, require aligned camera poses, or rely on additional regularizers. In this work, we introduce the first diffusion model able to perform fast, detailed reconstruction and generation of real-world 3D scenes. To achieve this, we make three contributions. First, we introduce a new neural scene representation, IB-planes, that can efficiently and accurately represent large 3D scenes, dynamically allocating more capacity as needed to capture details visible in each image. Second, we propose a denoising-diffusion framework to learn a prior over this novel 3D scene representation, using only 2D images without the need for any additional supervision signal such as masks or depths. This supports 3D reconstruction and generation in a unified architecture. Third, we develop a principled approach to avoid trivial 3D solutions when integrating the image-based rendering with the diffusion model, by dropping out representations of some images. We evaluate the model on several challenging datasets of real and synthetic images, and demonstrate superior results on generation, novel view synthesis and 3D reconstruction.