Research map / Video / Video generation
AnimateAnyMesh: A Feed-Forward 4D Foundation Model for Text-Driven Universal Mesh Animation
Zijie Wu, Chaohui Yu, Fan Wang, Xiang Bai
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 mesh adjacency attention mask in DyMeshVAE (ablation) | DyMeshVAE · DyMeshVAE reconstruction | reconstruction error: 0.500 -> 0.223 | Improved |
| add positional encoding for initial vertex features (ablation) | DyMeshVAE · DyMeshVAE reconstruction | reconstruction error: 0.443 -> 0.223 | Improved |
| add positional encoding for relative trajectory features (ablation) | DyMeshVAE · DyMeshVAE reconstruction | reconstruction error: 0.441 -> 0.223 | Improved |
| compute attention maps from initial frame features only (Sep Attn) (ablation) | DyMeshVAE · DyMeshVAE reconstruction | reconstruction error: 0.478 -> 0.223 | Improved |
| apply FPS to topology-aware features instead of raw coordinates (ablation) | DyMeshVAE · DyMeshVAE reconstruction | reconstruction error: 0.291 -> 0.223 | Improved |
| increase FPS sampling ratio from 1/32 to 1/4 for low-vertex mesh (369 vertices) (ablation) | DyMeshVAE · 369-vertex mesh | reconstruction error: 136.57 -> 8.10 | Improved |
| increase FPS sampling ratio from 1/8 to 1/4 for dense mesh (6,890 vertices) (ablation) | DyMeshVAE · 6,890-vertex mesh | reconstruction error: 0.51 -> 0.50 | No clear effect |
| increase max training vertex count from 4096 to 8192 (ablation) | AnimateAnyMesh · 10 mesh-prompt pairs | I2V: I2V 0.954 -> 0.985; Dy.Dg 0.693 -> 0.605 | Mixed |
| increase frames from 16 to 32 (ablation) | AnimateAnyMesh · 10 mesh-prompt pairs | Dynamic degree: 0.693 -> 0.737 | Improved |
| scale text-to-trajectory model from 200M to 740M parameters (ablation) | AnimateAnyMesh · 10 mesh-prompt pairs | I2V: 0.954 -> 0.968 | Improved |
| use feed-forward AnimateAnyMesh instead of optimization-based Animate3D | AnimateAnyMesh · Text-driven mesh animation | inference time: ~14min -> ~6s | Improved |
Later papers that compared with this method
4 later accepted papers report AnimateAnyMesh as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- BiMotion: B-spline Motion for Text-guided Dynamic 3D Character Generation (CVPR 2026 · better on 7 of 8 shared results)
- DiTex4D: Direct Text-Driven 4D Generation with Structured Latent Diffusion (ECCV 2026 · better on 3 of 4 shared results)
- AnimateAnyMesh++: A Flexible Feed-Forward Framework for High-Fidelity Text-Driven Mesh Animation (TPAMI 2026 · better on 4 of 4 shared results)
- RigMo: Unifying Rig and Motion Learning for Generative Animation (CVPR 2026 · better on 3 of 3 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
- Animate3D: Animating Any 3D Model with Multi-view Video Diffusion (NeurIPS 2024 · similarity 0.88)
- Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Models (CVPR 2024 · similarity 0.88)
- MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation (CVPR 2025 · similarity 0.87)
- Fast 4D Mesh Generation by Spatio-Temporal Attention Chains (NeurIPS 2026 · similarity 0.86)
- Progressive Rendering Distillation: Adapting Stable Diffusion for Instant Text-to-Mesh Generation without 3D Data (CVPR 2025 · similarity 0.86)
- Restage4D: Reanimating Deformable 3D Reconstruction from a Single Video (NeurIPS 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.
