Wonder3D: Single Image to 3D using Cross-Domain Diffusion

TL;DR

Wonder3D is a new method that converts single images into detailed 3D textured meshes efficiently by using a cross-domain diffusion model, multi-view attention, and normal fusion.

Contribution

It introduces a cross-domain diffusion approach with multi-view attention and a normal fusion algorithm for high-quality, consistent 3D reconstruction from single images.

Findings

Achieves high-quality 3D reconstructions from single images.

Demonstrates robust generalization across different images.

Offers improved efficiency over prior methods.

Abstract

In this work, we introduce Wonder3D, a novel method for efficiently generating high-fidelity textured meshes from single-view images.Recent methods based on Score Distillation Sampling (SDS) have shown the potential to recover 3D geometry from 2D diffusion priors, but they typically suffer from time-consuming per-shape optimization and inconsistent geometry. In contrast, certain works directly produce 3D information via fast network inferences, but their results are often of low quality and lack geometric details. To holistically improve the quality, consistency, and efficiency of image-to-3D tasks, we propose a cross-domain diffusion model that generates multi-view normal maps and the corresponding color images. To ensure consistency, we employ a multi-view cross-domain attention mechanism that facilitates information exchange across views and modalities. Lastly, we introduce a geometry-aware normal fusion algorithm that extracts high-quality surfaces from the multi-view 2D representations. Our extensive evaluations demonstrate that our method achieves high-quality reconstruction results, robust generalization, and reasonably good efficiency compared to prior works.