通过协同调控分组与特征提取，提升点云网络性能。
Unlocking Hidden Potential in Point Cloud Networks with Attention-Guided Grouping-Feature Coordination
- 设计轻量级模块协同优化分组与特征提取
- 在ModelNet40上达94.0%准确率，超越基线
- 适合追求高效点云分析的工程应用
点云分析已发展出多种网络架构，但现有工作多聚焦于结构创新。传统点基础架构通过顺序采样、分组和特征提取层处理原始点，存在未充分挖掘的潜力。我们发现，通过策略性模块集成可实现显著性能提升，而非依赖结构修改。本文提出分组-特征协同模块（GF-Core），一种轻量级可分离组件，同时调控分组层与特征提取层，实现更精细的特征聚合。此外，提出专为点输入设计的自监督预训练策略，增强模型在复杂点云分析场景下的鲁棒性。在ModelNet40数据集上，所提方法将基线网络提升至94.0%准确率，达到先进框架水平，同时保持架构简洁。在ScanObjectNN三个变体上，分别获得2.96%、6.34%和6.32%的提升。
原文摘要 · Abstract (English)
Point cloud analysis has evolved with diverse network architectures, while existing works predominantly focus on introducing novel structural designs. However, conventional point-based architectures - processing raw points through sequential sampling, grouping, and feature extraction layers - demonstrate underutilized potential. We notice that substantial performance gains can be unlocked through strategic module integration rather than structural modifications. In this paper, we propose the Grouping-Feature Coordination Module (GF-Core), a lightweight separable component that simultaneously regulates both grouping layer and feature extraction layer to enable more nuanced feature aggregation. Besides, we introduce a self-supervised pretraining strategy specifically tailored for point-based inputs to enhance model robustness in complex point cloud analysis scenarios. On ModelNet40 dataset, our method elevates baseline networks to 94.0% accuracy, matching advanced frameworks' performance while preserving architectural simplicity. On three variants of the ScanObjectNN dataset, we obtain improvements of 2.96%, 6.34%, and 6.32% respectively.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。