Research map / Point clouds / Point cloud classification
Point Cloud Self-supervised Learning via 3D to Multi-view Masked Learner
Zhimin Chen, Xuewei Chen, Xiao Guo, Yingwei Li, Longlong Jing, Liang Yang, Bing Li
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 token-level reconstruction (ablation) | Multiview-ML (Point-MAE base) · ScanObjectNN PB-T50-RS | accuracy (%): 85.18 -> 87.07 | Improved |
| add instance-level intra/inter-modality prediction on top of token reconstruction (ablation) | Multiview-ML · ScanObjectNN PB-T50-RS | accuracy (%): 87.07 -> 88.11 | Improved |
| add MSMH attention to token plus instance reconstruction (ablation) | Multiview-ML · ScanObjectNN PB-T50-RS | accuracy (%): 88.11 -> 88.93 | Improved |
| feed only 3D input rather than 3D plus 2D input (ablation) | Multiview-ML · ScanObjectNN PB-T50-RS | accuracy (%): 87.02 -> 88.93 | Improved |
| predict only 2D multi-view output instead of 3D and 2D (ablation) | Multiview-ML · ScanObjectNN PB-T50-RS | accuracy (%): 88.93 -> 84.25 | Hurt |
| pretrain with Multiview-ML before 3DETR fine-tuning | 3DETR · ScanNetV2 | AP50: 37.9 -> 43.3 | Improved |
| add Multiview-ML pretraining to Point-M2AE | Point-M2AE · ModelNet40 | accuracy (%): +0.4% | Improved |
| add 3D-to-multi-view projection and MSMH to I2P-MAE and ReCon | I2P-MAE, ReCon · ScanObjectNN | accuracy | 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.
- Point-M2AE — Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training (NeurIPS 2022) · better on 1 of 1 shared result
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
- Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training (NeurIPS 2022 · similarity 0.92)
- Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders (AAAI 2024 · similarity 0.92)
- 3D Feature Prediction for Masked-AutoEncoder-Based Point Cloud Pretraining (ICLR 2024 · similarity 0.91)
- MM-Point: Multi-View Information-Enhanced Multi-Modal Self-Supervised 3D Point Cloud Understanding (AAAI 2024 · similarity 0.90)
- RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning (AAAI 2025 · similarity 0.89)
- Multi-Scale Neighborhood Occupancy Masked Autoencoder for Self-Supervised Learning in LiDAR Point Clouds (CVPR 2025 · 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.
