Research map / Point clouds / Shape and surface completion
A Conditional Point Diffusion-Refinement Paradigm for 3D Point Cloud Completion
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 conditional DDPM to generate coarse point clouds and refine them with RFNet | PDR · MVP | EMD: 1.37 vs. 2.31 | Helped |
| Use conditional DDPM to generate coarse point clouds and refine them with RFNet | PDR · MVP-40 (50% missing) | EMD: 2.68 vs. 18.40 | Helped |
| Use conditional DDPM to generate coarse point clouds and refine them with RFNet | PDR · MVP-40 (12.5% missing) | EMD: 1.39 vs. 4.54 | Helped |
| Use conditional DDPM to generate coarse point clouds and refine them with RFNet | PDR · Completion3D | EMD: 1.75 vs. 2.69 | Helped |
| Add attention to PA-Deconv (ablation) | PDR CGNet · MVP, coarse generation | CD: 8.81 -> 8.71 | Helped |
| Add PA-Deconv to PointNet++ (ablation) | PDR CGNet · MVP, coarse generation | CD: 9.39 -> 8.81 | Helped |
| Use PA-Deconv and attention instead of PointNet++ (ablation) | PDR CGNet · MVP, coarse generation | CD: 9.39 -> 8.71 | Helped |
| Replace FT with concatenating noisy and partial clouds (ablation) | PDR CGNet · MVP, coarse generation | CD: 8.71 -> 10.79 | Hurt |
| Use only a global conditioner feature (ablation) | PDR CGNet · MVP, coarse generation | CD: 8.71 -> 11.99 | Hurt |
| Add attention to PA-Deconv (ablation) | PDR RFNet · MVP, refinement | CD: 5.96 -> 5.66 | Helped |
| Add PA-Deconv to PointNet++ (ablation) | PDR RFNet · MVP, refinement | CD: 6.03 -> 5.96 | Helped |
| Replace FT with concatenating noisy and partial clouds (ablation) | PDR RFNet · MVP, refinement | CD: 5.66 vs. 5.78 | Mixed |
Closest papers by meaning
- Digging into Intrinsic Contextual Information for High-fidelity 3D Point Cloud Completion (AAAI 2025 · similarity 0.90)
- PCDreamer: Point Cloud Completion Through Multi-view Diffusion Priors (CVPR 2025 · similarity 0.90)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.90)
- ComPC: Completing a 3D Point Cloud with 2D Diffusion Priors (ICLR 2025 · similarity 0.89)
- Position-Aware Guided Point Cloud Completion with CLIP Model (AAAI 2025 · similarity 0.89)
- PointInfinity: Resolution-Invariant Point Diffusion Models (CVPR 2024 · similarity 0.89)
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.