Direct and Explicit 3D Generation from a Single Image

TL;DR

This paper presents a fast, scalable method for generating detailed 3D models from a single image by combining multi-view 2D data with a modified diffusion model, improving quality and efficiency.

Contribution

The authors introduce a novel framework that directly produces explicit 3D surface geometry and textures from a single image using multi-view data and a modified Stable Diffusion model.

Findings

Outperforms existing methods in geometry and texture quality.

Achieves faster 3D generation times.

Produces high-resolution, detailed 3D models from a single image.

Abstract

Current image-to-3D approaches suffer from high computational costs and lack scalability for high-resolution outputs. In contrast, we introduce a novel framework to directly generate explicit surface geometry and texture using multi-view 2D depth and RGB images along with 3D Gaussian features using a repurposed Stable Diffusion model. We introduce a depth branch into U-Net for efficient and high quality multi-view, cross-domain generation and incorporate epipolar attention into the latent-to-pixel decoder for pixel-level multi-view consistency. By back-projecting the generated depth pixels into 3D space, we create a structured 3D representation that can be either rendered via Gaussian splatting or extracted to high-quality meshes, thereby leveraging additional novel view synthesis loss to further improve our performance. Extensive experiments demonstrate that our method surpasses existing baselines in geometry and texture quality while achieving significantly faster generation time.