Gen-3Diffusion: Realistic Image-to-3D Generation via 2D & 3D Diffusion Synergy

TL;DR

Gen-3Diffusion introduces a novel approach combining 2D and 3D diffusion models to generate realistic, multi-view consistent 3D objects and avatars from a single image, enhancing generalization and accuracy.

Contribution

The paper proposes a synchronized 2D and 3D diffusion framework that improves multi-view consistency and generalization in 3D object and avatar generation from single images.

Findings

Produces high-fidelity 3D objects and avatars

Enhances multi-view consistency in generated images

Demonstrates strong generalization to diverse shapes and clothing

Abstract

Creating realistic 3D objects and clothed avatars from a single RGB image is an attractive yet challenging problem. Due to its ill-posed nature, recent works leverage powerful prior from 2D diffusion models pretrained on large datasets. Although 2D diffusion models demonstrate strong generalization capability, they cannot guarantee the generated multi-view images are 3D consistent. In this paper, we propose Gen-3Diffusion: Realistic Image-to-3D Generation via 2D & 3D Diffusion Synergy. We leverage a pre-trained 2D diffusion model and a 3D diffusion model via our elegantly designed process that synchronizes two diffusion models at both training and sampling time. The synergy between the 2D and 3D diffusion models brings two major advantages: 1) 2D helps 3D in generalization: the pretrained 2D model has strong generalization ability to unseen images, providing strong shape priors for the 3D diffusion model; 2) 3D helps 2D in multi-view consistency: the 3D diffusion model enhances the 3D consistency of 2D multi-view sampling process, resulting in more accurate multi-view generation. We validate our idea through extensive experiments in image-based objects and clothed avatar generation tasks. Results show that our method generates realistic 3D objects and avatars with high-fidelity geometry and texture. Extensive ablations also validate our design choices and demonstrate the strong generalization ability to diverse clothing and compositional shapes. Our code and pretrained models will be publicly released on https://yuxuan-xue.com/gen-3diffusion.