Research map / Video / Video generation
AnimateAnyMesh++: A Flexible Feed-Forward Framework for High-Fidelity Text-Driven 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 |
|---|---|---|---|
| use chunk-based temporal compression instead of holistic compression (dim 256) (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0126 -> 0.0070 | Improved |
| chunk-based dim 256 vs holistic dim 1024 (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0083 -> 0.0070 | Improved |
| increase holistic channel dim from 256 to 1024 (ablation) | DyMeshVAE (no chunk) · 64-frame reconstruction | inference time: x1.00 -> x4.03 | Hurt |
| inject normals before positional encoding (Normal Enc1) (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0126 -> 0.0158 | Hurt |
| inject normals before neighborhood encoding (Normal Enc2) (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0126 -> 0.0115 | Improved |
| inject normals after neighborhood encoding (Normal Enc3) (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0126 -> 0.0118 | Improved |
| expand PLTA hops from L_hop=1 to 4 (ablation) | DyMeshVAE · 64-frame reconstruction | AVE: 0.0126 -> 0.0102 | Improved |
| increase hops from 4 to 5 (ablation) | PLTA · 64-frame reconstruction | AVE: 0.0102 -> 0.0104 | No clear effect |
| use learnable hop decay instead of fixed 0.5 (ablation) | PLTA · 64-frame reconstruction | AVE: 0.0102 -> 0.0115 | Hurt |
| add TDGW (ablation) | DyMeshVAE-Flex · 64-frame reconstruction | AVE: 0.0070 -> 0.0066 | Improved |
| add PLTA to Flex (ablation) | DyMeshVAE-Flex · DM_8192v_64f_test | rho2 abnormal edges: 0.0076 -> 0.0042 | Improved |
| add vertex normals to Flex + PLTA (ablation) | DyMeshVAE-Flex · DM_8192v_64f_test | AVE: 0.0063 -> 0.0054 | Improved |
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.
- AnimateAnyMesh — AnimateAnyMesh: A Feed-Forward 4D Foundation Model for Text-Driven Universal Mesh Animation (ICCV 2025) · 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
- Animate3D: Animating Any 3D Model with Multi-view Video Diffusion (NeurIPS 2024 · similarity 0.89)
- Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Models (CVPR 2024 · similarity 0.87)
- 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 0.87)
- How to Move Your Dragon: Text-to-Motion Synthesis for Large-Vocabulary Objects (ICML 2025 · similarity 0.86)
- Rigel3D: Rig-aware Latents for Animation-Ready 3D Asset Generation (NeurIPS 2026 · 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.
