Research map / 3D vision / Face and shape
Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic Flows
Kunal Gupta, Manmohan Chandraker
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 instance normalization layers (ablation) | NMF · ShapeNet auto-encoding | Chamfer-L2: 6.48 -> 5.00 | Improved |
| add instance normalization layers (ablation) | NMF · ShapeNet auto-encoding | Self-Intersection: 3.28 -> 0.03 | Improved |
| add instance normalization layers (ablation) | NMF · ShapeNet auto-encoding | NM-Faces: 2.94 -> 0.39 | Improved |
| add one refinement flow (0 to 1) (ablation) | NMF · ShapeNet auto-encoding | Chamfer-L2: 5.00 -> 4.93 | Improved |
| add second refinement flow (1 to 2) (ablation) | NMF · ShapeNet auto-encoding | Chamfer-L2: 4.93 -> 4.65 | Improved |
| add second refinement flow (1 to 2) (ablation) | NMF · ShapeNet auto-encoding | NM-Faces: 0.38 -> 0.73 | Hurt |
| add second refinement flow (1 to 2) (ablation) | NMF · ShapeNet auto-encoding | Time: 124 -> 189 | Hurt |
| NMF vs AtlasNet (auto-encoding) | NMF · ShapeNet auto-encoding | Self-Intersection: 24.80 -> 0.10 | Improved |
| NMF vs AtlasNet-O (auto-encoding) | NMF · ShapeNet auto-encoding | Chamfer-L2: 3.50 -> 5.54 | Hurt |
| NMF vs MeshRCNN (single-view reconstruction) | NMF · ShapeNet SVR | Chamfer-L2: 4.73 -> 7.82 | Hurt |
| NMF vs best baseline (single-view reconstruction) | NMF · ShapeNet SVR | Self-Intersection: 7.07 -> 0.12 | Improved |
| apply Laplacian smoothing to NMF output (SVR) | NMF · ShapeNet SVR | Self-Intersection: 0.12 -> 0.00 | 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.
- Pixel2Mesh — Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (ECCV 2018) · better on 3 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
- MeshXL: Neural Coordinate Field for Generative 3D Foundation Models (NeurIPS 2024 · similarity 0.87)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.86)
- Mosaic-SDF for 3D Generative Models (CVPR 2024 · similarity 0.86)
- Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models (NeurIPS 2026 · similarity 0.86)
- Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing (NeurIPS 2025 · similarity 0.86)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (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.
