Research map / Point clouds / Segmentation and detection
Learning Latent Part-Whole Hierarchies for Point Clouds
Xiang Gao, Wei Hu, Renjie Liao
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 control variate to REINFORCE estimator (ablation) | MC-REINFORCE · Chair middle level | accuracy: 72.39 ± 0.95 -> 73.75 ± 0.19 | Improved |
| add control variate to REINFORCE estimator (ablation) | MC-REINFORCE · Lamp middle level | accuracy: 75.65 ± 1.86 -> 76.23 ± 0.52 | Improved |
| add control variate to REINFORCE estimator (ablation) | MC-REINFORCE · Table middle level | accuracy: 70.40 ± 0.58 -> 71.64 ± 0.27 | Improved |
| add control variate to REINFORCE estimator (ablation) | MC-REINFORCE · Lamp top level | accuracy: 76.78 ± 0.82 -> 77.49 ± 0.59 | Improved |
| use latent part-whole model instead of PointNet | MC-REINFORCE · multi-level segmentation, middle level | mean accuracy: 59.53 -> 66.69 | Improved |
| use latent part-whole model instead of PointNet | MC-REINFORCE · multi-level segmentation, top level | mean accuracy: 85.19 -> 87.39 | Improved |
| use MC-REINFORCE inference instead of Pathwise | latent part-whole model · middle level | mean accuracy: 64.93 -> 66.69 | Improved |
| increase number of Monte Carlo samples (5 to 100) (ablation) | MC inference · middle level segmentation | accuracy and std | Improved |
| decrease temperature tau of pathwise estimator from large values (ablation) | MC-Pathwise · Chair, Lamp, Table middle level | accuracy | Mixed |
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.
- PointConv — PointConv: Deep Convolutional Networks on 3D Point Clouds (CVPR 2019) · better on 2 of 2 shared results
- PointCNN — PointCNN: Convolution On X-Transformed Points (NeurIPS 2018) · better on 2 of 2 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
- Clustering-based Supervised Training for 3D Perception (TPAMI 2026 · similarity 0.90)
- S$^2$AM3D: Scale-controllable Part Segmentation of 3D Point Clouds (CVPR 2026 · similarity 0.90)
- PointGLR: Unsupervised Structural Representation Learning of 3D Point Clouds (TPAMI 2022 · similarity 0.90)
- Self-Supervised Learning on 3D Point Clouds by Learning Discrete Generative Models (CVPR 2021 · similarity 0.89)
- Fully-Convolutional Point Networks for Large-Scale Point Clouds (ECCV 2018 · similarity 0.89)
- Multi-Modality Affinity Inference for Weakly Supervised 3D Semantic Segmentation (AAAI 2024 · similarity 0.88)
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.
