Research map / 3D vision / Face and shape
TextMesh4D: Zero-shot Text-to-4D Mesh Generation
Sisi Dai, Xinxin Su, Kai Xu
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 |
|---|---|---|---|
| replace vertex-level deformation with face-level Jacobian Deformation Field (ablation) | TextMesh4D · GPT-4V ablation | Overall (%): 1.1 -> 77.3 (vs Var-A); 3.9 -> 77.3 (vs Var-B) | Improved |
| add global semantic regularizer (G-SR) (ablation) | TextMesh4D · GPT-4V ablation | Overall (%): 7.4 -> 77.3 | Improved |
| add local ARAP regularizer (L-SR) (ablation) | TextMesh4D · GPT-4V ablation | Overall (%): 10.3 -> 77.3 | Improved |
| add rigging structure to vertex deformation (Var-B vs Var-A) (ablation) | vertex-based deformation · GPT-4V ablation | Structure quality (%): 1.0 -> 7.3 | Improved |
| use TextMesh4D (mesh) instead of NeRF/3DGS 4D baselines | text-to-4D generation · Text-to-4D prompts | CLIP score: 31.93 (best baseline STAG4D) -> 32.32 | Improved |
| use TextMesh4D on a single 24GB GPU instead of A100 80GB clusters | text-to-4D generation · GPU requirement | hardware: 1 * A5000 24GB vs 128 * A100 80GB (AYG) | Improved |
| fixed mesh connectivity for topology-changing prompts | TextMesh4D · failure cases | qualitative | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Puppeteer — Puppeteer: Rig and Animate Your 3D Models (NeurIPS 2025) · better on 2 of 2 shared results
- 4D-fy — 4D-fy: Text-to-4D Generation Using Hybrid Score Distillation Sampling (CVPR 2024) · better on 2 of 2 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
- Geometry Image Diffusion: Fast and Data-Efficient Text-to-3D with Image-Based Surface Representation (ICLR 2025 · similarity 0.87)
- Diffusion4D: Fast Spatial-temporal Consistent 4D generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.87)
- Fast 4D Mesh Generation by Spatio-Temporal Attention Chains (NeurIPS 2026 · similarity 0.87)
- A Unified Approach for Text- and Image-guided 4D Scene Generation (CVPR 2024 · similarity 0.86)
- DreamMesh4D: Video-to-4D Generation with Sparse-Controlled Gaussian-Mesh Hybrid Representation (NeurIPS 2024 · similarity 0.86)
- Sculpt3D: Multi-View Consistent Text-to-3D Generation with Sparse 3D Prior (CVPR 2024 · 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.
