Research map / 3D vision / Face and shape
RADmesh: Remesh-Aware Mesh Deformation
Nam Anh Dinh, Itai Lang, Oded Stein, Rana Hanocka
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 the proposed remesh schedule to MeshUp | MeshUp · 64 shape-prompt pairs | VQA score x100: 66.093 -> 69.360 | Improved |
| add the proposed remesh schedule to MeshUp | MeshUp · 64 shape-prompt pairs | face quality x100: 84.831 -> 96.310 | Improved |
| use full RADmesh (scale parameter, state interpolation) instead of MeshUp with our remesh | MeshUp + our remesh · 64 shape-prompt pairs | VQA score x100: 69.360 -> 72.930 | Improved |
| use RADmesh instead of MeshUp + our remesh for face budget | RADmesh · 64 shape-prompt pairs | average face count: 29461.34 -> 16963.75 | Improved |
| add optimizable per-vertex scale parameter (ablation) | RADmesh · local growths (qualitative) | growth extent and detail | Improved |
| interpolate optimizer state through remeshing (ablation) | RADmesh · qualitative stability and face count | stability, symmetry, face count | Improved |
| use coarse-to-fine remesh schedule instead of one fine remesh at the start (ablation) | RADmesh · local growths (qualitative) | growth quality | Improved |
| use uniform isotropic remeshing instead of curvature-adaptive remeshing (ablation) | RADmesh · local growths (qualitative) | growth extent | Improved |
| add initial normal inflation of selected region (ablation) | RADmesh · local growths (qualitative) | growth onset | Improved |
| add leaky ReLU floor of 0.98 on scale (ablation) | RADmesh · global deformation (qualitative) | shrinkage | Improved |
| use RADmesh deformation quantity instead of Continuous Remeshing / Large Steps with CSD loss | inverse-rendering remeshers · text-guided generative deformation (qualitative) | formation of new parts | Improved |
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.
- Instant3dit — Instant3dit: Multiview Inpainting for Fast Editing of 3D Objects (CVPR 2025) · better on 4 of 4 shared results
- Geometry in Style — Geometry in Style: 3D Stylization via Surface Normal Deformation (CVPR 2025) · 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
- Few-Shot Physically-Aware Articulated Mesh Generation via Hierarchical Deformation (ICCV 2023 · similarity 0.88)
- Deep Feature Deformation Weights (CVPR 2026 · similarity 0.87)
- ARMesh: Autoregressive Mesh Generation via Next-Level-of-Detail Prediction (NeurIPS 2025 · similarity 0.87)
- Geometry in Style: 3D Stylization via Surface Normal Deformation (CVPR 2025 · similarity 0.86)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (CVPR 2024 · similarity 0.86)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.85)
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.
