Research map / Point clouds / Shape and surface completion
GRNet: Gridding Residual Network for Dense Point Cloud Completion
Haozhe Xie, Hongxun Yao, Shangchen Zhou, Jiageng Mao, Shengping Zhang, Wenxiu Sun
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 |
|---|---|---|---|
| replace prior methods with GRNet | best prior (PCN) · ShapeNet | Chamfer Distance overall (x10^-4): 4.016 -> 2.723 | Improved |
| use GRNet | MSN · ShapeNet | F-Score@1% overall: 0.705 -> 0.708 | Improved |
| use GRNet | TopNet · Completion3D | Chamfer Distance overall (x10^-4): 14.25 -> 10.64 | Improved |
| use GRNet | TopNet · KITTI | Consistency (x10^-3): 0.568 -> 0.313 | Improved |
| use GRNet | MSN · KITTI | Fidelity Distance (x10^3): 0.434 -> 0.816 | Hurt |
| increase Gridding resolution from 32^3 to 64^3 (ablation) | GRNet · ShapeNet | Chamfer Distance complete (x10^-4): 5.943 -> 2.723 | Improved |
| increase Gridding resolution from 64^3 to 128^3 (ablation) | GRNet · ShapeNet | Chamfer Distance complete (x10^-4): 2.723 -> 2.732 | No clear effect |
| add Cubic Feature Sampling with three feature maps (ablation) | GRNet · ShapeNet | Chamfer Distance (x10^-4): 11.375 -> 2.723 | Improved |
| add more feature maps into Cubic Feature Sampling (one to three) (ablation) | GRNet · ShapeNet | Chamfer Distance (x10^-4): 2.922 -> 2.723 | Improved |
| add Gridding Loss (ablation) | GRNet · ShapeNet | Chamfer Distance complete (x10^-4): 4.460 -> 2.723 | Improved |
| add Gridding Loss (ablation) | GRNet · ShapeNet | F-Score@1% complete: 0.624 -> 0.708 | Improved |
| increase Gridding Loss resolution from 64^3 to 128^3 (ablation) | GRNet · ShapeNet | Chamfer Distance complete (x10^-4): 3.427 -> 2.723 | Improved |
Later papers that compared with this method
25 later accepted papers report GRNet as a baseline in their results tables; 21 of them report a better number on most of the results they share with it.
- SVDFormer: Complementing Point Cloud via Self-view Augmentation and Self-structure Dual-generator (ICCV 2023 · better on 2 of 2 shared results)
- AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers (TPAMI 2023 · better on 2 of 2 shared results)
- CP3: Unifying Point Cloud Completion by Pretrain-Prompt-Predict Paradigm (TPAMI 2023 · better on 2 of 2 shared results)
- Variational Relational Point Completion Network for Robust 3D Classification (TPAMI 2023 · better on 2 of 2 shared results)
- Parametric Surface Constrained Upsampler Network for Point Cloud (AAAI 2023 · better on 0 of 1 shared result)
- VAPCNet: Viewpoint-Aware 3D Point Cloud Completion (ICCV 2023 · better on 1 of 1 shared result)
- PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-Step Point Moving Paths (TPAMI 2022 · better on 4 of 4 shared results)
- SeedFormer: Patch Seeds Based Point Cloud Completion with Upsample Transformer (ECCV 2022 · better on 3 of 3 shared results)
17 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FoldingNet — FoldingNet: Point Cloud Auto-Encoder via Deep Grid Deformation (CVPR 2018) · better on 4 of 5 shared results
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
- SoftPool++: An Encoder–Decoder Network for Point Cloud Completion (IJCV 2022 · similarity 0.90)
- Learning Geometric Transformation for Point Cloud Completion (IJCV 2023 · similarity 0.90)
- Point Cloud Completion by Skip-Attention Network With Hierarchical Folding (CVPR 2020 · similarity 0.90)
- A Conditional Point Diffusion-Refinement Paradigm for 3D Point Cloud Completion (ICLR 2022 · similarity 0.90)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.89)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.89)
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.
