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

By Xiaoxiao Long, Yuanchen Guo, Cheng Lin, Yuan Liu, Zhiyang Dou, Lingjie Liu, Yuexin Ma, Song-Hai Zhang, Marc Habermann, C. Theobalt and 1 more. Computer Vision and Pattern Recognition, 2023

825 citations recorded by the source corpus.

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 di-rectly 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 single-view reconstruction tasks, we pro-pose a cross-domain diffusion model that generates multi-view normal maps and the corresponding color images. To ensure the consistency of generation, 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 in only 2 r-;» 3 minutes. Our extensive evaluations demonstrate that our method achieves high-quality reconstruction results, robust generalization, and good efficiency compared to prior works.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/d2c5565a039f464b778e0f2263da418ef42e98b0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (2023, methodology): “After extracting the 3D mesh from the denoised volume, we further enhance the texture by employing a lightweight optimization with multi-view images as supervision.”
ImageDream: Image-Prompt Multi-view Diffusion for 3D Generation (2023, methodology): “We also note recent parallel developments, such as Won-der3D [21], which incorporate normal diffused outputs into original diffusion models, and DreamCraft3D [40], which employ a second-stage DreamBooth-like model fine-tuning for enhanced texture modeling.”
LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation (2024, methodology): “During inference, our method leverages existing image or text to multi-view diffusion models [27, 43, 44, 51] to produce multi-view images as inputs for our Gaussian fusion network.”
SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusion (2024, background): “MVDream [45], Consistent123 [53], and Wonder3D [27] also train multi-view diffusion models, yet still require post-processing for video rendering.”
GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models (2023, neutral): “Some single-image-to-3D meth-ods [13, 19, 30, 33–36, 38, 54, 60, 65, 66, 71, 78, 79, 82, 83] also employ similar ideas.”
Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation (2025, neutral): “Most works force the multi-view generated latents to communicate with each other by manipulating the self-attention layers with 3D-aware masks [36, 48, 88, 60, 95, 79, 78, 55].”
InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Models (2024, methodology): “A series of works [24, 26, 43, 48, 51] further propose multi-view generation models, thereby the optimization processes can be guided by multiple novel views simultaneously.”
GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting (2023, positive): “4.3, it takes approximately 3 minutes to generate a 3D mesh using the method from [25] and an additional 2 minutes to transfer this mesh into 3D Gaussians and reﬁne it, while the composition process of two Gaussians takes less than 1 second.”
GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image (2024, methodology): “Inspired by [35], we leverage a geometry switcher to enable a single stable diffusion model to generate depth or normal through indicators.”
GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation (2024, background): “LGM [92], One-2-3-45 [54], One-2-3-45++ [53], and Wonder3D [58] also combine multiview diffusion and reconstruction for single image-to-3D generation, but they produce blurrier texture and noticeable geometry artifacts.”