Muses: Designing, Composing, Generating Nonexistent Fantasy 3D Creatures without Training

2026-01-06

arXiv.org

AI Summary

Existing methods for generating fantastical 3D creatures often rely on training data, part-based optimization, or 2D image guidance, frequently resulting in structurally incoherent or visually distorted outputs. This work proposes a training-free, feedforward generation framework that constructs anatomically plausible 3D skeletons through graph-constrained reasoning, followed by voxel-level latent space assembly and skeleton-conditioned image-guided texture modeling. The approach achieves high-quality 3D creature synthesis with coherent morphology and consistent visual style. To the best of our knowledge, this is the first method to enable fully training-free creation of imaginative 3D creatures, attaining state-of-the-art performance in both visual fidelity and alignment with textual descriptions, while also supporting flexible 3D editing.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

We present Muses, the first training-free method for fantastic 3D creature generation in a feed-forward paradigm. Previous methods, which rely on part-aware optimization, manual assembly, or 2D image generation, often produce unrealistic or incoherent 3D assets due to the challenges of intricate part-level manipulation and limited out-of-domain generation. In contrast, Muses leverages the 3D skeleton, a fundamental representation of biological forms, to explicitly and rationally compose diverse elements. This skeletal foundation formalizes 3D content creation as a structure-aware pipeline of design, composition, and generation. Muses begins by constructing a creatively composed 3D skeleton with coherent layout and scale through graph-constrained reasoning. This skeleton then guides a voxel-based assembly process within a structured latent space, integrating regions from different objects. Finally, image-guided appearance modeling under skeletal conditions is applied to generate a style-consistent and harmonious texture for the assembled shape. Extensive experiments establish Muses'state-of-the-art performance in terms of visual fidelity and alignment with textual descriptions, and potential on flexible 3D object editing. Project page: https://luhexiao.github.io/Muses.github.io/.

Problem

• fantasy 3D creature generation
• 3D asset coherence
• out-of-domain generation
• part-level manipulation
• structure-aware 3D creation

Innovation

• training-free
• 3D skeleton
• structure-aware generation
• voxel-based assembly
• image-guided texturing