Research map / Point clouds / Point cloud classification
Efficient Spiking Point Mamba for Point Cloud Analysis
Peixi Wu, Bosong Chai, Menghua Zheng, Wei Li, Zhangchi Hu, Jie Chen, Zheyu Zhang, Hebei Li, Xiaoyan Sun
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 SNN-adapted Gate II and bidirectional SSM (SMB) over vanilla Mamba block (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 88.3 -> 90.2 | Improved |
| add SNN-adapted Gate II and bidirectional SSM (SMB) over vanilla Mamba block (ablation) | SPM · ModelNet40 | overall accuracy: 91.4 -> 92.3 | Improved |
| add bidirectional SSM branch (Gate I) (ablation) | SPM Model I · ScanObjectNN PB-T50-RS | overall accuracy: 83.1 -> 83.8 | Improved |
| add SNN-adapted Gate II (ablation) | SPM Model I · ScanObjectNN PB-T50-RS | overall accuracy: 83.1 -> 83.5 | Improved |
| flip time dimension instead of no flip in SSM branch (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 88.9 -> 90.2 | Improved |
| add Token Flip rather than Time Flip (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 90.2 -> 89.6 | Hurt |
| apply both Token Flip and Time Flip (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 90.2 -> 89.2 | Hurt |
| add HDE finite forward sliding and infinite backward extension (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 88.6 -> 90.2 | Improved |
| add HDE finite forward sliding and infinite backward extension (ablation) | SPM · ModelNet40 | overall accuracy: 91.5 -> 92.3 | Improved |
| increase time step from 1 to 4 (ablation) | SPM · ScanObjectNN PB-T50-RS | overall accuracy: 83.3 -> 84.2 | Improved |
| increase time step from 4 to 6 (ablation) | SPM · ScanObjectNN PB-T50-RS | overall accuracy: 84.2 -> 84.3 | No clear effect |
| replace LIF neuron with ILIF neuron (ablation) | SPM · ScanObjectNN OBJ-BG | overall accuracy: 90.2 -> 91.5 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.
- PointMamba — PointMamba: A Simple State Space Model for Point Cloud Analysis (NeurIPS 2024) · better on 0 of 4 shared results
- Point-M2AE — Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training (NeurIPS 2022) · better on 1 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
- Point-to-Spike Residual Learning for Energy-Efficient 3D Point Cloud Classification (AAAI 2024 · similarity 0.89)
- Spiking Point Transformer for Point Cloud Classification (AAAI 2025 · similarity 0.89)
- Efficient 3D Recognition with Event-driven Spike Sparse Convolution (AAAI 2025 · similarity 0.88)
- Noise-Injected Spiking Graph Convolution for Energy-Efficient 3D Point Cloud Denoising (AAAI 2025 · similarity 0.87)
- Point Cloud Mamba: Point Cloud Learning via State Space Model (AAAI 2025 · similarity 0.86)
- Rethinking Spiking Neural Networks from an Ensemble Learning Perspective (ICLR 2025 · 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.
