Research map / Point clouds / Shape and surface completion
Learning a Structured Latent Space for Unsupervised Point Cloud Completion
Check 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 |
|---|---|---|---|
| learn a unified latent space with structured latent supervision | unsupervised point cloud completion · CRN average | CD (×10⁴): 14.9 -> 12.2 | Helped |
| learn a unified latent space with structured latent supervision | unsupervised point cloud completion · CRN average | F1: 83.9 -> 85.6 | Helped |
| use unified-space element-wise multiplication rather than code concatenation (ablation) | Uni. Space · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 19.1 -> 18.3 | Helped |
| use unified-space element-wise multiplication rather than code addition (ablation) | Uni. Space · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 18.6 -> 18.3 | Helped |
| add triplet ranking supervision (ablation) | Uni. Space · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 18.3 -> 17.2 | Helped |
| add N-pair ranking supervision (ablation) | Uni. Space · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 18.3 -> 15.4 | Helped |
| add the point cloud discriminator (ablation) | Full Model · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 19.4 -> 15.4 | Helped |
| add the latent code discriminator (ablation) | Full Model · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 20.8 -> 15.4 | Helped |
| add latent code swapping between related partial shapes (ablation) | Full Model · CRN average (Chair, Lamp, Sofa, Table) | CD (×10⁴): 15.4 -> 17.7 | Helped |
| combine the proposed method with GAN inversion | Ours · real scans | UCD (×10⁴): 3.2 -> 1.1; 2.7 -> 0.87; 3.3 -> 1.1; 2.7 -> 0.87; 4.2 -> 0.76 | Helped |
Closest papers by meaning
- Point Cloud Completion by Skip-Attention Network With Hierarchical Folding (CVPR 2020 · similarity 0.90)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.89)
- Digging into Intrinsic Contextual Information for High-fidelity 3D Point Cloud Completion (AAAI 2025 · similarity 0.89)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.89)
- Position-Aware Guided Point Cloud Completion with CLIP Model (AAAI 2025 · similarity 0.88)
- Orthogonal Dictionary Guided Shape Completion Network for Point Cloud (AAAI 2024 · 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.