Research map / Point clouds / Point cloud classification
3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions
Dong Wook Shu, Sung Woo Park, Junseok Kwon
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 r-GAN with tree-GAN | r-GAN (dense) · ShapeNet Chair | COV-CD: 33 -> 58 | Improved |
| replace r-GAN with tree-GAN | r-GAN (dense) · ShapeNet Chair | FPD: 1.860 -> 0.809 | Improved |
| replace r-GAN with tree-GAN | r-GAN (dense) · ShapeNet Airplane | FPD: 1.016 -> 0.439 | Improved |
| replace r-GAN with tree-GAN | r-GAN (dense) · ShapeNet (all 16 classes) | FPD: 4.726 -> 3.600 | Improved |
| replace r-GAN with tree-GAN | r-GAN (dense) · ShapeNet (all 16 classes) | JSD: 0.171 -> 0.105 | Improved |
| replace r-GAN with tree-GAN | r-GAN (dense) · ShapeNet (all 16 classes) | COV-EMD: 29 -> 39 | Improved |
| replace Valsesia et al. (up.) with tree-GAN | Valsesia et al. (up.) · ShapeNet Chair | JSD: 0.100 -> 0.119 | Hurt |
| replace Valsesia et al. (up.) with tree-GAN | Valsesia et al. (up.) · ShapeNet Airplane | JSD: 0.083 -> 0.097 | Hurt |
| replace Valsesia et al. (up.) with tree-GAN | Valsesia et al. (up.) · ShapeNet Airplane | COV-CD: 31 -> 61 | Improved |
| replace Valsesia et al. (up.) with tree-GAN | Valsesia et al. (up.) · ShapeNet Chair | MMD-CD: 0.0029 -> 0.0016 | Improved |
| use different branching degrees such as {1,2,2,2,2,2,64}, {1,2,4,16,4,2,2}, {1,32,4,2,2,2,2} (ablation) | tree-GAN · ShapeNet | FPD convergence | No clear effect |
| select different ancestors in the tree (ablation) | tree-GAN generator · ShapeNet Airplane | semantic part generation (qualitative) | Improved |
Closest papers by meaning
- Learning Localized Generative Models for 3D Point Clouds via Graph Convolution (ICLR 2019 · similarity 0.90)
- CPCGAN: A Controllable 3D Point Cloud Generative Adversarial Network with Semantic Label Generating (AAAI 2021 · similarity 0.89)
- PT2PC: Learning to Generate 3D Point Cloud Shapes from Part Tree Conditions (ECCV 2020 · similarity 0.88)
- Learning Progressive Point Embeddings for 3D Point Cloud Generation (CVPR 2021 · similarity 0.88)
- Generative PointNet: Deep Energy-Based Learning on Unordered Point Sets for 3D Generation, Reconstruction and Classification (CVPR 2021 · similarity 0.87)
- PU-GAN: A Point Cloud Upsampling Adversarial Network (ICCV 2019 · similarity 0.87)
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.
