Research map / 3D vision / Face and shape
Learning Conjugate Direction Fields for Planar Quadrilateral Mesh Generation
Jiong Tao, Yong-Liang Yang, Bailin Deng
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 direction smoothness loss Lds (ablation) | CDF network · Test set ablation | number of singularities: 7.02 -> 4.91 | Improved |
| add direction smoothness loss Lds (ablation) | CDF network · Test set ablation | stroke inconsistency delta: 7.38 -> 8.31 | Mixed |
| add stroke consistency loss Ldc (ablation) | CDF network · Test set ablation | stroke inconsistency delta: 10.32 -> 8.31 | Improved |
| add stroke consistency loss Ldc (ablation) | CDF network · Test set ablation | number of singularities: 4.67 -> 4.91 | No clear effect |
| use context-aware stroke feature instead of PCT (ablation) | CDF network · Test set ablation | stroke inconsistency delta: 20.98 -> 8.31 | Improved |
| use context-aware stroke feature instead of PCT (ablation) | CDF network · Test set ablation | CDF angle to ground truth theta: 19.06 -> 11.30 | Improved |
| replace optimization-based CDF generation with learned network | CDF generation · Surface CDF generation time | time (s): 40.412s -> 0.571s (Face) | Improved |
| vertex perturbation optimization after quadrangulation | PQ mesh from learned CDF · Test set (300 models) | mean planarity error: 0.0067 -> 0.0023 | Improved |
Closest papers by meaning
- PoNQ: a Neural QEM-based Mesh Representation (CVPR 2024 · similarity 0.84)
- Mesh BDF: Barycentric Dominance Field for 3D Native Mesh Generation (NeurIPS 2026 · similarity 0.82)
- Polyhedral Complex Derivation from Piecewise Trilinear Networks (NeurIPS 2024 · similarity 0.82)
- High Resolution UDF Meshing via Iterative Networks (NeurIPS 2025 · similarity 0.82)
- An Intrinsic Vector Heat Network (ICML 2024 · similarity 0.82)
- Geometry-Informed Neural Networks (ICML 2025 · similarity 0.81)
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.
