ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation

TL;DR

ProlificDreamer introduces a variational score distillation framework for text-to-3D generation, improving diversity and quality over traditional score distillation sampling by modeling 3D parameters as random variables.

Contribution

The paper proposes VSD, a novel particle-based variational framework that generalizes SDS, enhancing diversity and fidelity in text-to-3D generation with high-resolution outputs.

Findings

VSD outperforms SDS across various CFG weights.

ProlificDreamer generates high-fidelity 3D models with complex effects.

The approach achieves detailed, photo-realistic NeRF and meshes.

Abstract

Score distillation sampling (SDS) has shown great promise in text-to-3D generation by distilling pretrained large-scale text-to-image diffusion models, but suffers from over-saturation, over-smoothing, and low-diversity problems. In this work, we propose to model the 3D parameter as a random variable instead of a constant as in SDS and present variational score distillation (VSD), a principled particle-based variational framework to explain and address the aforementioned issues in text-to-3D generation. We show that SDS is a special case of VSD and leads to poor samples with both small and large CFG weights. In comparison, VSD works well with various CFG weights as ancestral sampling from diffusion models and simultaneously improves the diversity and sample quality with a common CFG weight (i.e., $7.5$). We further present various improvements in the design space for text-to-3D such as distillation time schedule and density initialization, which are orthogonal to the distillation algorithm yet not well explored. Our overall approach, dubbed ProlificDreamer, can generate high rendering resolution (i.e., $512\times512$) and high-fidelity NeRF with rich structure and complex effects (e.g., smoke and drops). Further, initialized from NeRF, meshes fine-tuned by VSD are meticulously detailed and photo-realistic. Project page and codes: https://ml.cs.tsinghua.edu.cn/prolificdreamer/