Research map / Point clouds / Shape and surface completion
Patch-Based Progressive 3D Point Set Upsampling
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 |
|---|---|---|---|
| Use the multi-stage architecture rather than iterating one 2× model (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.69 -> 0.54 | Helped |
| Train the units end-to-end rather than separately (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.73 -> 0.54 | Helped |
| Train units progressively rather than simultaneously (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.55 -> 0.54 | Helped |
| Add dense feature extraction (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.61 -> 0.54 | Helped |
| Use code-assignment feature expansion instead of the PU-Net counterpart (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.73 -> 0.54 | Helped |
| Add inter-level skip connections (ablation) | Patch-based progressive upsampling network · Sketchfab, 625-point input, 16× upsampling | CD: 0.61 -> 0.54 | Helped |
| Use patch-based progressive upsampling instead of the strongest baseline | 16× point set upsampling · Sketchfab, 625-point input | CD: 0.67 -> 0.54 | Helped |
| Use patch-based progressive upsampling instead of the strongest baseline | 16× point set upsampling · Sketchfab, 5000-point input | CD: 0.09 -> 0.06 | Helped |
| Use adaptive patch extraction across progressive stages (ablation) | Multi-stage PU-Net on MNIST-CP · MNIST-CP | Geometric reconstruction quality | Helped |
| Train multi-stage units end-to-end rather than separately (ablation) | Multi-stage PU-Net on MNIST-CP · MNIST-CP | Geometric reconstruction quality | Helped |
Closest papers by meaning
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.88)
- CUF: Continuous Upsampling Filters (CVPR 2023 · similarity 0.87)
- SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network (AAAI 2025 · similarity 0.86)
- SPU-PMD: Self-Supervised Point Cloud Upsampling via Progressive Mesh Deformation (CVPR 2024 · similarity 0.84)
- Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2024 · similarity 0.84)
- HashPoint: Accelerated Point Searching and Sampling for Neural Rendering (CVPR 2024 · similarity 0.84)
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.