Controllable Text-to-3D Generation via Surface-Aligned Gaussian Splatting

TL;DR

This paper introduces a controllable text-to-3D generation framework that combines a novel neural network architecture with an efficient multi-stage pipeline, enabling high-quality, guided 3D content creation using surface-aligned Gaussian splatting.

Contribution

The work presents MVControl, a new neural network architecture for integrating various input conditions into multi-view diffusion models, and a multi-stage 3D generation pipeline using Gaussian representations and hybrid guidance.

Findings

Achieves robust generalization in controllable 3D generation.

Enables fine-grained geometry editing on 3D meshes.

Demonstrates high-quality 3D content creation guided by diverse conditions.

Abstract

While text-to-3D and image-to-3D generation tasks have received considerable attention, one important but under-explored field between them is controllable text-to-3D generation, which we mainly focus on in this work. To address this task, 1) we introduce Multi-view ControlNet (MVControl), a novel neural network architecture designed to enhance existing pre-trained multi-view diffusion models by integrating additional input conditions, such as edge, depth, normal, and scribble maps. Our innovation lies in the introduction of a conditioning module that controls the base diffusion model using both local and global embeddings, which are computed from the input condition images and camera poses. Once trained, MVControl is able to offer 3D diffusion guidance for optimization-based 3D generation. And, 2) we propose an efficient multi-stage 3D generation pipeline that leverages the benefits of recent large reconstruction models and score distillation algorithm. Building upon our MVControl architecture, we employ a unique hybrid diffusion guidance method to direct the optimization process. In pursuit of efficiency, we adopt 3D Gaussians as our representation instead of the commonly used implicit representations. We also pioneer the use of SuGaR, a hybrid representation that binds Gaussians to mesh triangle faces. This approach alleviates the issue of poor geometry in 3D Gaussians and enables the direct sculpting of fine-grained geometry on the mesh. Extensive experiments demonstrate that our method achieves robust generalization and enables the controllable generation of high-quality 3D content. Project page: https://lizhiqi49.github.io/MVControl/.