Our model is built upon pre-trained 2D stable diffusion models [45] to leverage its strong generalization.

We propose a multi-view cross-domain diffusion scheme, which operates on two distinct domains to generate multi-view consistent normal maps and color images.

We propose that the distribution of 3d assets can be modeled as a joint distribution of its corresponding 2d multi-view normal maps and corresponding images.

Diffusion models [22, 52] are first proposed to gradually recover images from a specifically designed degradation process

Instead of performing a time-consuming per-shape optimization guided by 2D diffusion models, some works attempt to directly train 3D diffusion models

Recent compelling successes in 2D diffusion models and large vision language models provide new possibilities for generating 3d assets.

In this work, we introduce Wonder3D, a novel method for efficiently generating high-fidelity textured meshes from single-view images.

To evaluate the quality of the single-view reconstructions, we adopt two commonly used metrics Chamfer Distances (CD) and Volume IoU between ground-truth shapes

We adopt Zero123 [31], RealFusion [38], Magic123 [44], One-2-3-45 [30], Point-E [41], Shap-E [25] and a recent work SyncDreamer [33] as baseline methods.

We train our model on the LVIS subset of the Objaverse dataset [9], which comprises approximately 30,000+ objects following a cleanup process.