Research map / Point clouds / Point cloud classification
PointGAC: Geometric-Aware Codebook for Masked Point Modeling
Abiao Li, Chenlei Lv, Yuming Fang, Yifan Zuo, Jian Zhang, Guofeng Mei
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 online k-means codebook alone to the regression baseline (ablation) | Point2Vec baseline · ModelNet40 (w/o vote) | accuracy: 92.5 -> 92.1 | Hurt |
| add codebook maintenance mechanism to online k-means codebook (ablation) | PointGAC · ModelNet40 (w/o vote) | accuracy: 92.1 -> 93.7 | Improved |
| add geometry-aware partitioning to baseline (ablation) | Point2Vec baseline · ModelNet40 (w/o vote) | accuracy: 92.5 -> 92.8 | Improved |
| add geometry-aware partitioning to codebook model (ablation) | PointGAC · ModelNet40 (w/o vote) | accuracy: 93.7 -> 94.3 | Improved |
| geometry-aware partitioning versus KNN at mask ratio 0.8 (ablation) | PointGAC · ModelNet40 (w/o vote) | accuracy: 93.7 -> 94.3 | Improved |
| increase mask ratio from 0.8 to 0.9 (ablation) | PointGAC · ModelNet40 (w/o vote) | accuracy: 94.3 -> 93.6 | Hurt |
| use meaningful perturbation instead of random perturbation of codebook (ablation) | online k-means codebook · ModelNet40 (w/o vote) | accuracy: 93.4 -> 94.3 | Improved |
| add random perturbation to online k-means codebook (ablation) | online k-means codebook · ModelNet40 (w/o vote) | accuracy: 92.7 -> 93.4 | Improved |
| replace Sinkhorn-based uniform-assignment codebook with online k-means plus meaningful perturbation (ablation) | codebook construction · ModelNet40 (w/o vote) | accuracy: 93.1 -> 94.3 | Improved |
| replace regression-based teacher feature alignment with cluster-assignment alignment | masked point modeling · ModelNet40 (linear SVM) | accuracy: 92.1 -> 93.5 | Improved |
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.
- RIMAE — RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning (AAAI 2025) · 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
- A Linear Time and Space Local Point Cloud Geometry Encoder via Vectorized Kernel Mixture (VecKM) (ICML 2024 · similarity 0.89)
- LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modeling (NeurIPS 2024 · similarity 0.89)
- PointCFormer: A Relation-Based Progressive Feature Extraction Network for Point Cloud Completion (AAAI 2025 · similarity 0.87)
- RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning (AAAI 2025 · similarity 0.87)
- 3D Feature Prediction for Masked-AutoEncoder-Based Point Cloud Pretraining (ICLR 2024 · similarity 0.86)
- PCP-MAE: Learning to Predict Centers for Point Masked Autoencoders (NeurIPS 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.
