Research map / Point clouds / Shape and surface completion
Surface Representation for Point Clouds
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 |
|---|---|---|---|
| Add Umbrella RepSurf to PointNet++ (SSG) with multi-scale inference | PointNet++ (SSG) · ModelNet40 | OA: 94.2 -> 94.7 | Helped |
| Add Umbrella RepSurf to PointNet++ (SSG) with multi-scale inference | PointNet++ (SSG) · ScanObjectNN PB_T50_RS | OA: 82.8 -> 84.6 | Helped |
| Add Umbrella RepSurf to PointNet++ (SSG) | PointNet++ (SSG) · S3DIS 6-fold | mIoU: 73.5 -> 74.3 | Helped |
| Add Umbrella RepSurf to PointNet++ (SSG) | PointNet++ (SSG) · S3DIS Area-5 | mIoU: 70.4 -> 68.9 | Mixed |
| Replace GroupFree PointNet++ backbone with RepSurf-U | GroupFree⁶,²⁵⁶ · ScanNetV2 detection | mAP@0.25: 67.3 -> 68.8 | Helped |
| Use post-computed rather than pre-computed Umbrella RepSurf (ablation) | Umbrella RepSurf · ModelNet40 | accuracy: 93.06 -> 93.90 | Helped |
| Add surface position and random inverse to post-computed Umbrella RepSurf (ablation) | Umbrella RepSurf · ModelNet40 | accuracy: 93.90 -> 94.46 | Helped |
| Add spherical polar auxiliary (ablation) | Umbrella RepSurf · ModelNet40 | accuracy: 93.97 -> 94.46 | Helped |
| Use cylindrical instead of spherical polar auxiliary (ablation) | Umbrella RepSurf · ModelNet40 | accuracy: 94.46 -> 93.89 | Hurt |
| Apply channel de-differentiation before rather than after the linear function (ablation) | PointNet++ · ModelNet40 | accuracy: 94.08 -> 92.70 | Hurt |
| Increase Umbrella RepSurf group size from k=8 to k=16 (ablation) | Umbrella RepSurf · ModelNet40 | accuracy: 94.46 -> 94.32 | No clear effect |
Closest papers by meaning
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (CVPR 2025 · similarity 0.86)
- Digging into Intrinsic Contextual Information for High-fidelity 3D Point Cloud Completion (AAAI 2025 · similarity 0.86)
- A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functions (CVPR 2025 · similarity 0.86)
- X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition (CVPR 2024 · similarity 0.85)
- Learning Bijective Surface Parameterization for Inferring Signed Distance Functions from Sparse Point Clouds with Grid Deformation (CVPR 2025 · similarity 0.85)
- RepKPU: Point Cloud Upsampling with Kernel Point Representation and Deformation (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.