Research map / Point clouds / Shape and surface completion
PUNO: A Neural Operator Framework for Point Cloud Upsampling
Zijian Xiao, Yining Xu, Yingjie Huang, Li Yao
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 |
|---|---|---|---|
| add vertex displacement network before manifold mapping (ablation) | PUNO · ShapeNet 16x | P2F avg: 7.892 -> 2.04 | Improved |
| add vertex displacement network (ablation) | PUNO · ShapeNet 16x | CD: 0.615 -> 0.524 | Improved |
| add neural kernel integral operator on top of data-space fitting (ablation) | PUNO · ShapeNet 16x | P2F avg: 3.632 -> 2.04 | Improved |
| use global Galerkin attention instead of kNN local operator (ablation) | PUNO · ShapeNet 16x | HD: 6.192 -> 5.446 | Improved |
| use Galerkin attention instead of softmax attention (ablation) | PUNO · inference memory at 8x | GiB: 27.6 GiB -> 2.0 GiB | Improved |
| use quadratic instead of linear patch mapping (ablation) | PUNO · ShapeNet 16x | P2F avg: 2.498 -> 2.04 | Improved |
| use quadratic instead of cubic patch mapping (ablation) | PUNO · ShapeNet 16x | P2F avg: 2.732 -> 2.04 | Improved |
| use PUNO instead of RepKPU | point cloud upsampling · ShapeNet 16x | P2F avg: 1.275 vs 2.04 | Hurt |
| use PUNO instead of RepKPU | point cloud upsampling · ShapeNet 16x | CD: 0.592 -> 0.524 | Improved |
| use PUNO instead of baselines under Gaussian noise std 0.01 | point cloud upsampling · ShapeNet noisy | P2F mean: 6.199 -> 4.004 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- RepKPU — RepKPU: Point Cloud Upsampling with Kernel Point Representation and Deformation (CVPR 2024) · better on 4 of 6 shared results
- SPU-IMR — SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network (AAAI 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PUGeo-Net: A Geometry-centric Network for 3D Point Cloud Upsampling (ECCV 2020 · similarity 0.88)
- SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network (AAAI 2025 · similarity 0.87)
- Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2024 · similarity 0.87)
- PLATYPUS: Progressive Local Surface Estimator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2025 · similarity 0.86)
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.86)
- TULIP: Transformer for Upsampling of LiDAR Point Clouds (CVPR 2024 · 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.
