Research map / 3D vision / Face and shape
Hybrid Vector-Occupancy Field for Robust Implicit 3D Surface Reconstruction
Yue Wu, Zhigang Gao, Tengfei Xiao, Can Qin, Yongzhe Yuan, Hao Li, Kaiyuan Feng, Wenping Ma
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 |
|---|---|---|---|
| use hybrid occupancy-guided vector field instead of occupancy field only (ablation) | HvoF · ABC dataset | EMD (x1e-2): 2.182 -> 1.137 | Improved |
| use hybrid field instead of occupancy field only (ablation) | HvoF · ABC dataset | F-score @1e-4 (x1e-2): 98.547 -> 98.753 | Improved |
| use hybrid field instead of occupancy field only (ablation) | HvoF · ABC dataset | Chamfer-L1 CD (x1e-2): 0.234 -> 0.266 | Hurt |
| use hybrid occupancy-truncated field instead of untruncated vector field (ablation) | HvoF · ABC dataset | Chamfer-L1 CD (x1e-2): 0.349 -> 0.266 | Improved |
| use hybrid field instead of untruncated vector field (ablation) | HvoF · ABC dataset | EMD (x1e-2): 2.391 -> 1.137 | Improved |
| use hybrid field instead of untruncated vector field (ablation) | HvoF · ABC dataset | F-score @1e-4 (x1e-2): 93.916 -> 98.753 | Improved |
| shell thickness r=0.10 instead of r=0.05 (ablation) | HvoF · ABC dataset | Chamfer-L1 CD (x1e-2): 0.279 -> 0.266 | Improved |
| shell thickness r=0.10 instead of r=0.15 (ablation) | HvoF · ABC dataset | Chamfer-L1 CD (x1e-2): 0.339 -> 0.266 | Improved |
| shell thickness r=0.10 instead of r=0.15 (ablation) | HvoF · ABC dataset | F-score @1e-4 (x1e-2): 96.009 -> 98.753 | Improved |
| shell thickness r=0.10 instead of r=0.05 (ablation) | HvoF · ABC dataset | EMD (x1e-2): 1.316 -> 1.137 | Improved |
| replace NVF with HvoF on ShapeNet base | NVF · ShapeNet base categories (seen) | Chamfer-L1 CD (x1e-2): 0.452 -> 0.299 | Improved |
| replace NVF with HvoF on MGN cross-domain | NVF · MGN garments (cross-domain) | Chamfer-L1 CD (x1e-2): 0.311 -> 0.310 | No clear effect |
Closest papers by meaning
- Neural Vector Fields: Implicit Representation by Explicit Learning (CVPR 2023 · similarity 0.89)
- Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks and Zero-Curl Regularization (TPAMI 2025 · similarity 0.88)
- Unsupervised Occupancy Learning from Sparse Point Cloud (CVPR 2024 · similarity 0.87)
- Details Enhancement in Unsigned Distance Field Learning for High-fidelity 3D Surface Reconstruction (AAAI 2025 · similarity 0.86)
- Learning 3D Equivariant Implicit Function with Patch-Level Pose-Invariant Representation (NeurIPS 2024 · similarity 0.86)
- A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functions (CVPR 2025 · 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.
