Research map / 3D vision / 3D generation
Muses: Designing, Composing, Generating Nonexistent Fantasy 3D Creatures without Training
Hexiao Lu, Xiaokun Sun, Zeyu Cai, Hao Guo, Ying Tai, Jian Yang, Zhenyu Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add LLM-guided assembly reasoning (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA1 score: 0.6967 -> 0.9254 | Improved |
| add LLM-guided assembly reasoning (ablation) | Muses · Fantasy creature prompts (30 samples) | CLIP score: 0.2573 -> 0.2878 | Improved |
| add skinning-weight region mapping (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA1 score: 0.7090 -> 0.9254 | Improved |
| add skinning-weight region mapping (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA2 score: 0.7081 -> 0.8496 | Improved |
| interpolate in 16^3 voxel space rather than 64^3 SLAT space (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA1 score: 0.7326 -> 0.9254 | Improved |
| add geometrically invariant texture editing (ablation) | Muses · Fantasy creature prompts (30 samples) | CLIP score: 0.2532 -> 0.2878 | Improved |
| add geometrically invariant texture editing (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA1 score: 0.7990 -> 0.9254 | Improved |
| add style-consistent texture generation (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA1 score: 0.8359 -> 0.9254 | Improved |
| add style-consistent texture generation (ablation) | Muses · Fantasy creature prompts (30 samples) | VQA2 score: 0.7902 -> 0.8496 | Improved |
| Muses compared with OmniPart (manually assembled) | Muses vs OmniPart · Fantasy creature prompts (30 samples) | Text alignment (user study): 9.84 -> 85.40 | Improved |
| Muses compared with Trellis-Text-to-3D | Muses vs Trellis-Text-to-3D · Fantasy creature prompts (30 samples) | VQA score: 0.7565 -> 0.9254 | Improved |
| increase skeleton complexity from easy to hard | LLM-guided assembly · Skeleton complexity bins (100 cases each) | LLM reasoning failure rate: 4% -> 11% | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GaussianDreamer — GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models (CVPR 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Muses: 3D-Controllable Image Generation via Multi-Modal Agent Collaboration (AAAI 2025 · similarity 0.87)
- Rigel3D: Rig-aware Latents for Animation-Ready 3D Asset Generation (NeurIPS 2026 · similarity 0.87)
- ShapeCraft: LLM Agents for Structured, Textured and Interactive 3D Modeling (NeurIPS 2025 · similarity 0.87)
- Puppeteer: Rig and Animate Your 3D Models (NeurIPS 2025 · similarity 0.86)
- SKDream: Controllable Multi-view and 3D Generation with Arbitrary Skeletons (CVPR 2025 · similarity 0.86)
- How to Move Your Dragon: Text-to-Motion Synthesis for Large-Vocabulary Objects (ICML 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
