Research map / 3D vision / Face and shape
GetMesh: A Controllable Model for High-quality Mesh Generation and Manipulation
Ben Fei, Jinyi Wang, Lei Bai, K. J. Ray Liu, Xudong Xu, Weidong Yang, Ya Zhang, Ying He, Dahua Lin, Zhaoyang Lyu, Bo Dai
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 32-64 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet Airplane | Shading-FID: 32.10 -> 206.57 | Hurt |
| use 2048-4096 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet Airplane | Shading-FID: 32.10 -> 237.36 | Hurt |
| use 512-1024 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet | Shading-FID: 32.10 -> 44.32 (Airplane), 121.42 -> 113.22 (Car) | Mixed |
| use 2048-4096 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet Airplane | Coverage (CD) percent: 47.6 -> 2.4 | Hurt |
| add refinement module to mesh autoencoder (ablation) | GetMesh autoencoder · ShapeNet test | Mask IoU: 0.961 -> 0.968 | Improved |
| add refinement module to mesh autoencoder (ablation) | GetMesh autoencoder · ShapeNet test | Chamfer distance x1e-3: 1.46 -> 1.34 | Improved |
| use 512-1024 latent points in autoencoder instead of 128-256 (ablation) | GetMesh autoencoder · ShapeNet test | Chamfer distance x1e-3: 1.34 -> 1.07 | Improved |
| use 512-1024 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet generation | seconds per sample: 1.17 -> 8.14 | Hurt |
| use 2048-4096 latent points instead of 128-256 (ablation) | GetMesh · ShapeNet generation | seconds per sample: 1.17 -> 92.18 | Hurt |
| use compact latent points for diffusion instead of 512 vectors | GetMesh vs 3DShape2VecSet · ShapeNet generation | seconds per sample: 6.56 -> 1.17 | Improved |
| use GetMesh instead of 3DShape2VecSet | multi-category generation · ShapeNet | Shading-FID: 23.90 -> 16.50 (Table), 101.38 -> 121.42 (Car) | Mixed |
| use GetMesh instead of 3DShape2VecSet | multi-category generation · ShapeNet Chair | 1-NNA (CD) percent: 60.66 -> 52.05 | 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.
- 3DILG — 3DILG: Irregular Latent Grids for 3D Generative Modeling (NeurIPS 2022) · 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
- PivotMesh: Generic 3D Mesh Generation via Pivot Vertices Guidance (ICLR 2025 · similarity 0.89)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.87)
- ARMesh: Autoregressive Mesh Generation via Next-Level-of-Detail Prediction (NeurIPS 2025 · similarity 0.87)
- MeshArt: Generating Articulated Meshes with Structure-Guided Transformers (CVPR 2025 · similarity 0.86)
- CraftsMan3D: High-fidelity Mesh Generation with 3D Native Diffusion and Interactive Geometry Refiner (CVPR 2025 · similarity 0.86)
- TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing (CVPR 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.
