提出新自监督方法Sonata，让点云模型在少量数据下也能高效准确地完成3D任务。
Sonata: Self-Supervised Learning of Reliable Point Representations
- 通过隐藏空间信息和增强特征依赖，避免点云表示退化为低级几何特征。
- 仅用1%数据就将线性探测准确率提升至72.5%，比之前方法高出近一倍。
- 适合追求高效3D感知的科研与工业应用，尤其适用于数据稀缺场景。
本文探讨在数据有限、计算资源少的情况下，是否存在可靠的3D自监督点云模型，可通过简单线性探测应用于多种3D任务。我们发现现有方法在线性探测评估中表现不佳，根源在于所谓的‘几何捷径’——导致表示退化为低级空间特征。这一挑战源于点云数据稀疏性，是3D独有的问题。为此，我们提出两种策略：遮蔽空间信息并增强对输入特征的依赖，最终通过自蒸馏构建了包含140k点云的Sonata模型。该方法简洁直观，学习到的表示强且可靠：零样本可视化展现语义分组，且通过最近邻关系体现良好空间推理能力。Sonata在扫描数据集ScanNet上将线性探测准确率从21.8%提升至72.5%（提升三倍），仅用1%数据即接近先前方法全量数据性能。全微调后，在室内与室外3D感知任务中均达到当前最优水平。
原文摘要 · Abstract (English)
In this paper, we question whether we have a reliable self-supervised point cloud model that can be used for diverse 3D tasks via simple linear probing, even with limited data and minimal computation. We find that existing 3D self-supervised learning approaches fall short when evaluated on representation quality through linear probing. We hypothesize that this is due to what we term the "geometric shortcut", which causes representations to collapse to low-level spatial features. This challenge is unique to 3D and arises from the sparse nature of point cloud data. We address it through two key strategies: obscuring spatial information and enhancing the reliance on input features, ultimately composing a Sonata of 140k point clouds through self-distillation. Sonata is simple and intuitive, yet its learned representations are strong and reliable: zero-shot visualizations demonstrate semantic grouping, alongside strong spatial reasoning through nearest-neighbor relationships. Sonata demonstrates exceptional parameter and data efficiency, tripling linear probing accuracy (from 21.8% to 72.5%) on ScanNet and nearly doubling performance with only 1% of the data compared to previous approaches. Full fine-tuning further advances SOTA across both 3D indoor and outdoor perception tasks.
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。