Research map / Point clouds / Shape and surface completion
HSDF: Hybrid Sign and Distance Field for Modeling Surfaces with Arbitrary Topologies
Li Wang, Jie Yang, Weikai Chen, Xiaoxu Meng, Bo Yang, Jintao Li, Lin Gao
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 |
|---|---|---|---|
| replace NDF (UDF + Ball-Pivoting) with HSDF-Net + masked Marching Cubes | HSDF-Net · Meshing 256^3 | meshing time per model: 780m -> 95s | Improved |
| replace NDF with HSDF-Net + Marching Cubes | HSDF-Net · Meshing 128^3 | meshing time per model: 58m -> 17s | Improved |
| replace NDF with HSDF-Net + Marching Cubes | HSDF-Net · Meshing 64^3 | meshing time per model: 89s -> 5s | Improved |
| replace NDF with HSDF-Net | HSDF-Net · Meshing 256^3 | storage per model: 1276M -> 10M | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Complex shapes mean, 3K points | Chamfer distance (x1e-4): 0.242 -> 0.223 | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Complex shapes mean, 10K points | Chamfer distance (x1e-4): 0.121 -> 0.128 | Hurt |
| use HSDF-Net instead of NDF | HSDF-Net · Complex shapes mean, 3K points | Oriented normal consistency: 0.381 -> 0.757 | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Complex shapes mean, 10K points | Oriented normal consistency: 0.462 -> 0.836 | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Complex shapes mean, 3K points | Normal consistency: 0.868 -> 0.892 | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Lamp, 10K points | Chamfer distance (x1e-4): 0.142 -> 0.166 | Hurt |
| use HSDF-Net instead of NDF | HSDF-Net · Watertight Cars (DISN), 3000 points | Chamfer-L2 (x1e-4): 0.191 -> 0.155 | Improved |
| use HSDF-Net instead of NDF | HSDF-Net · Watertight Cars (DISN), 300 points | Chamfer-L2 (x1e-4): 1.51 -> 1.34 | Improved |
Closest papers by meaning
- DUDF: Differentiable Unsigned Distance Fields with Hyperbolic Scaling (CVPR 2024 · similarity 0.90)
- High Resolution UDF Meshing via Iterative Networks (NeurIPS 2025 · similarity 0.90)
- MIND: Material Interface Generation from UDFs for Non-Manifold Surface Reconstruction (NeurIPS 2025 · similarity 0.89)
- Unsigned Orthogonal Distance Fields: An Accurate Neural Implicit Representation for Diverse 3D Shapes (CVPR 2024 · similarity 0.89)
- Details Enhancement in Unsigned Distance Field Learning for High-fidelity 3D Surface Reconstruction (AAAI 2025 · similarity 0.89)
- A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functions (CVPR 2025 · 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.
