DreamBeast: Distilling 3D Fantastical Animals with Part-Aware Knowledge Transfer

TL;DR

DreamBeast introduces a novel part-aware knowledge transfer method that leverages stable diffusion models to efficiently generate high-quality, part-specific 3D fantastical animals, overcoming previous limitations in speed and part-level understanding.

Contribution

It proposes a new part-aware knowledge transfer mechanism that improves 3D animal generation by integrating part-level semantics from diffusion models into a 3D representation.

Findings

Enhanced 3D animal quality with detailed part control

Reduced computational overhead compared to existing methods

Effective part-level semantic understanding in 3D generation

Abstract

We present DreamBeast, a novel method based on score distillation sampling (SDS) for generating fantastical 3D animal assets composed of distinct parts. Existing SDS methods often struggle with this generation task due to a limited understanding of part-level semantics in text-to-image diffusion models. While recent diffusion models, such as Stable Diffusion 3, demonstrate a better part-level understanding, they are prohibitively slow and exhibit other common problems associated with single-view diffusion models. DreamBeast overcomes this limitation through a novel part-aware knowledge transfer mechanism. For each generated asset, we efficiently extract part-level knowledge from the Stable Diffusion 3 model into a 3D Part-Affinity implicit representation. This enables us to instantly generate Part-Affinity maps from arbitrary camera views, which we then use to modulate the guidance of a multi-view diffusion model during SDS to create 3D assets of fantastical animals. DreamBeast significantly enhances the quality of generated 3D creatures with user-specified part compositions while reducing computational overhead, as demonstrated by extensive quantitative and qualitative evaluations.