Research map / Point clouds / Shape and surface completion
Symmetric Shape-Preserving Autoencoder for Unsupervised Real Scene 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 USSPA instead of Unpaired | Unsupervised completion on the USSPA dataset, single-category training · USSPA dataset, 10 categories (AVG) | L1 Chamfer Distance: 10.47 -> 8.56 | Helped |
| Use USSPA instead of Unpaired | Unsupervised completion · USSPA dataset, multi-category training | L1 Chamfer Distance: 10.12 -> 8.76 | Helped |
| Use classifier-guided instead of probability-guided discriminators | USSPA · USSPA dataset, multi-category training | L1 Chamfer Distance: 8.96 -> 8.76 | Helped |
| Train one model across categories rather than separately | USSPA · USSPA dataset | L1 Chamfer Distance: 8.96 -> 8.76 | No clear effect |
| Use USSPA instead of Unpaired | Unsupervised completion trained on ScanNet · PCN Dataset (AVG) | L1 Chamfer Distance × 10²: 14.87 -> 12.63 | Helped |
| Add the symmetry learning module (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 16.43 -> 18.43 | Helped |
| Add the refinement autoencoder (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 17.08 -> 18.43 | Helped |
| Add local feature grouping (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 15.72 -> 18.43 | Helped |
| Add the feature discriminator (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 17.11 -> 18.43 | Helped |
| Add the point discriminator (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 15.68 -> 18.43 | Helped |
| Add single-direction Chamfer Distance (ablation) | USSPA · USSPA dataset, chair | F-score at 0.1%: 17.84 -> 18.43 | Helped |
| Add upsampling refinement (ablation) | USSPA · USSPA dataset, chair, 512 points | F-score at 0.1%: 5.60 -> 6.46 | Helped |
Closest papers by meaning
- Learning a Structured Latent Space for Unsupervised Point Cloud Completion (CVPR 2022 · similarity 0.89)
- Unsupervised Point Cloud Completion and Segmentation by Generative Adversarial Autoencoding Network (NeurIPS 2022 · similarity 0.89)
- Unsupervised Learning of Probably Symmetric Deformable 3D Objects From Images in the Wild (CVPR 2020 · similarity 0.88)
- Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders (AAAI 2024 · similarity 0.87)
- Adaptive Siamese Masked Autoencoder with Global Optimization for Unsupervised Point Cloud Shape Correspondence (AAAI 2025 · similarity 0.87)
- PCDreamer: Point Cloud Completion Through Multi-view Diffusion Priors (CVPR 2025 · similarity 0.87)
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.