Research map / Point clouds / Shape and surface completion
Progressive Point Cloud Deconvolution Generation Network
Le Hui, Rui Xu, Jin Xie, Jianjun Qian, Jian Yang
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 |
|---|---|---|---|
| use learning-based bilateral interpolation instead of reshape (ablation) | PDGN · ShapeNet Chair | JSD: 8.69 -> 1.71 | Improved |
| use learning-based bilateral interpolation instead of bilinear interpolation (ablation) | PDGN · ShapeNet Chair | JSD: 5.02 -> 1.71 | Improved |
| use bilateral instead of bilinear interpolation (ablation) | PDGN · ShapeNet Chair | 1-NNA-CD: 69.23 -> 52.38 | Improved |
| use shape-preserving loss instead of plain adversarial loss (ablation) | PDGN · ShapeNet Chair | JSD: 3.28 -> 1.71 | Improved |
| use shape-preserving loss instead of EMD loss between resolutions (ablation) | PDGN · ShapeNet Chair | JSD: 3.35 -> 1.71 | Improved |
| use shape-preserving loss instead of CD loss between resolutions (ablation) | PDGN · ShapeNet Chair | JSD: 3.34 -> 1.71 | Improved |
| generate 2048 points progressively instead of stopping at 256 points (ablation) | PDGN · ShapeNet Chair | JSD: 5.57 -> 1.71 | Improved |
| generate 2048 points instead of 1024 points (ablation) | PDGN · ShapeNet Chair | JSD: 2.18 -> 1.71 | Improved |
| generate 1024 points instead of 512 points (ablation) | PDGN · ShapeNet Chair | MMD-EMD: 9.67 -> 11.0 | Hurt |
| set k around 20 in the neighborhood (ablation) | PDGN · ShapeNet Chair | generation metrics | Improved |
| use PDGN instead of PointFlow | PDGN vs PointFlow · ShapeNet Airplane | COV-CD: 46.91 -> 64.98 | Improved |
| use PDGN instead of PointFlow on Car | PDGN vs PointFlow · ShapeNet Car | MMD-CD: 0.91 -> 1.07 | No clear effect |
Closest papers by meaning
- Learning Progressive Point Embeddings for 3D Point Cloud Generation (CVPR 2021 · similarity 0.90)
- Interpolated Convolutional Networks for 3D Point Cloud Understanding (ICCV 2019 · similarity 0.87)
- MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image (AAAI 2019 · similarity 0.87)
- TopNet: Structural Point Cloud Decoder (CVPR 2019 · similarity 0.87)
- Learning to Train a Point Cloud Reconstruction Network without Matching (ECCV 2022 · similarity 0.87)
- CRA-PCN: Point Cloud Completion with Intra- and Inter-level Cross-Resolution Transformers (AAAI 2024 · similarity 0.86)
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.
