Research map / 3D vision / Face and shape
MeshRipple: Structured Autoregressive Generation of Artist-Meshes
Junkai Lin, Hang Long, Huipeng Guo, Jielei Zhang, Jiayi Yang, Tianle Guo, Yang Yang, Jianwen Li, Wenxiao ZHANG, Matthias Nießner, Wei Yang
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 frontier mask (ablation) | MeshRipple · Ablation (5k-face meshes) | CD: 52.67 vs 54.79 without (+2.12) | Improved |
| add context injection (ablation) | MeshRipple · Ablation (5k-face meshes) | CD: 52.67 vs 60.56 without (+7.89) | Improved |
| add non-manifold encoding (ablation) | MeshRipple · Ablation (5k-face meshes) | CD: 52.67 vs 57.75 without (+5.08) | Improved |
| add root constraint (expansive prediction) (ablation) | MeshRipple · Ablation (5k-face meshes) | CD: 52.67 vs 65.08 without (+12.41) | Improved |
| add root constraint (expansive prediction) (ablation) | MeshRipple · Ablation (5k-face meshes) | HD: 0.1058 vs 0.1180 without (+0.0122) | Improved |
| add NSCA (ablation) | MeshRipple · Ablation (5k-face meshes) | CD: 52.67 vs 52.43 without (-0.24) | No clear effect |
| add NSCA (ablation) | MeshRipple · Ablation (5k-face meshes) | HD: 0.1058 vs 0.1025 without (-0.0033) | Hurt |
| add NSCA (ablation) | MeshRipple · Memory at 100K faces, 2K window | GPU memory (GB): 62.26 -> 31.42 | Improved |
| add NSCA (ablation) | MeshRipple · Runtime at 100K faces, 2K window | time (s): 0.984 -> 0.458 | Improved |
| predict root as index offset instead of coordinates (ablation) | MeshRipple · Root prediction ablation | MMD: 19.26 -> 15.26 | Improved |
| predict root as index offset instead of coordinates (ablation) | MeshRipple · Root prediction ablation | COV: 47.56 -> 58.29 | Improved |
| increase truncation window from 1k to 2k (ablation) | MeshRipple · Dense meshes (160) | CD: 0.051948 -> 0.050651 | 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.
- DeepMesh — DeepMesh: Differentiable Iso-Surface Extraction (TPAMI 2024) · better on 5 of 6 shared results
- MeshAnythingV2 — MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 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
- ARMesh: Autoregressive Mesh Generation via Next-Level-of-Detail Prediction (NeurIPS 2025 · similarity 0.91)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.91)
- TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing (CVPR 2025 · similarity 0.89)
- Scaling Mesh Generation via Compressive Tokenization (CVPR 2025 · similarity 0.89)
- MeshArt: Generating Articulated Meshes with Structure-Guided Transformers (CVPR 2025 · similarity 0.88)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (CVPR 2024 · similarity 0.88)
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.
