Research map / Point clouds / Shape and surface completion
Flexible-Weighted Chamfer Distance: Enhanced Objective Function for Point Cloud Completion
Jie Li, Shengwei Tian, Long Yu, Xin Ning
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 CD-l2 with static FCD | AdaPoinTr · ShapeNet55 | avg DCD: 0.613 -> 0.537 | Improved |
| replace CD-l2 with static FCD | AdaPoinTr · ShapeNet55 | avg EMD: 28.983 -> 29.253 | Hurt |
| replace CD-l2 with static FCD | SeedFormer · ShapeNet55 | avg DCD: 0.596 -> 0.528 | Improved |
| replace CD-l2 with static FCD | SeedFormer · ShapeNet55 | avg F-Score: 0.412 -> 0.403 | Hurt |
| replace CD with static FCD | AdaPoinTr · PCN | EMD: 23.790 -> 21.395 | Improved |
| replace CD with static FCD | AdaPoinTr · PCN | CD-l1: 6.509 -> 6.591 | Hurt |
| replace CD with static FCD | SeedFormer · PCN | CD-l1: 6.617 -> 6.711 | Hurt |
| use DCD as standalone loss instead of CD (ablation) | AdaPoinTr · PCN | F-Score@1%: 0.847 -> 0.827 | Hurt |
| use linear schedule instead of static weights (ablation) | AdaPoinTr + FCD · PCN | CD-l1: 6.591 -> 6.538 | Improved |
| use uncertainty weighting instead of static weights (ablation) | AdaPoinTr + FCD · PCN | DCD: 0.514 -> 0.520 | Mixed |
| add FCD for real scans | AdaPoinTr · KITTI | Consistency: 0.569 -> 0.525 | Improved |
| replace CD with static FCD | RepKPU · PU-GAN 16x | EMD: 47.363 -> 46.591 | Improved |
Closest papers by meaning
- Balanced Chamfer Distance as a Comprehensive Metric for Point Cloud Completion (NeurIPS 2021 · similarity 0.94)
- SymmCompletion: High-Fidelity and High-Consistency Point Cloud Completion with Symmetry Guidance (AAAI 2025 · similarity 0.88)
- CMG-Net: Robust Normal Estimation for Point Clouds via Chamfer Normal Distance and Multi-Scale Geometry (AAAI 2024 · similarity 0.88)
- Fully Dynamic Algorithms for Chamfer Distance (NeurIPS 2025 · similarity 0.87)
- Position-Aware Guided Point Cloud Completion with CLIP Model (AAAI 2025 · similarity 0.85)
- Beyond Chamfer Distance: Granular Order-aware Evaluation Metric For Online Mapping (NeurIPS 2026 · similarity 0.85)
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.
