ExactDreamer: High-Fidelity Text-to-3D Content Creation via Exact Score Matching

TL;DR

ExactDreamer introduces Exact Score Matching, a novel method that enhances high-fidelity text-to-3D content creation by ensuring exact recovery in the diffusion process, overcoming over-smoothing and reconstruction errors of previous models.

Contribution

It proposes Exact Score Matching (ESM) with auxiliary variables and LoRA to improve detail and accuracy in text-to-3D generation, addressing limitations of prior diffusion-based methods.

Findings

ESM outperforms existing methods in detailed 3D generation

Demonstrates superior fidelity and content preservation

Effective in handling complex text prompts

Abstract

Text-to-3D content creation is a rapidly evolving research area. Given the scarcity of 3D data, current approaches often adapt pre-trained 2D diffusion models for 3D synthesis. Among these approaches, Score Distillation Sampling (SDS) has been widely adopted. However, the issue of over-smoothing poses a significant limitation on the high-fidelity generation of 3D models. To address this challenge, LucidDreamer replaces the Denoising Diffusion Probabilistic Model (DDPM) in SDS with the Denoising Diffusion Implicit Model (DDIM) to construct Interval Score Matching (ISM). However, ISM inevitably inherits inconsistencies from DDIM, causing reconstruction errors during the DDIM inversion process. This results in poor performance in the detailed generation of 3D objects and loss of content. To alleviate these problems, we propose a novel method named Exact Score Matching (ESM). Specifically, ESM leverages auxiliary variables to mathematically guarantee exact recovery in the DDIM reverse process. Furthermore, to effectively capture the dynamic changes of the original and auxiliary variables, the LoRA of a pre-trained diffusion model implements these exact paths. Extensive experiments demonstrate the effectiveness of ESM in text-to-3D generation, particularly highlighting its superiority in detailed generation.