用知识蒸馏让小模型高效分割3D点云，兼顾几何与语义信息。
SRKD: Towards Efficient 3D Point Cloud Segmentation via Structure- and Relation-aware Knowledge Distillation
- 通过点间相似性匹配，传递教师模型的结构依赖关系。
- 跨样本构建小批量，提升学生模型对通用几何结构的感知能力。
- 在<15M参数下达到顶尖性能，适合实际部署场景。
3D点云分割因大规模Transformer模型计算复杂和部署受限而面临实际挑战。为此，我们提出一种新型结构与关系感知的知识蒸馏框架SRKD，将大尺寸冻结教师模型（>100M参数）中的丰富几何与语义知识迁移到轻量级学生模型（<15M参数）。具体而言，我们设计基于亲和矩阵的关系对齐模块，通过点级相似性匹配，将教师模型的结构依赖传递给学生，增强其学习上下文交互的能力；同时引入跨样本小批量构建策略，使学生能够感知稳定且泛化的几何结构，而非局限于单个样本内部。此外，采用KL散度对齐语义分布，并结合真实标签监督以进一步提升分割精度。该方法在显著降低模型复杂度的同时实现当前最优性能，验证了其在真实部署场景中的有效性与高效性。代码已开源：https://github.com/itsnotacie/SRKD。
原文摘要 · Abstract (English)
3D point cloud segmentation faces practical challenges due to the computational complexity and deployment limitations of large-scale transformer-based models. To address this, we propose a novel Structure- and Relation-aware Knowledge Distillation framework, named SRKD, that transfers rich geometric and semantic knowledge from a large frozen teacher model (>100M) to a lightweight student model (<15M). Specifically, we propose an affinity matrix-based relation alignment module, which distills structural dependencies from the teacher to the student through point-wise similarity matching, enhancing the student's capability to learn contextual interactions. Meanwhile, we introduce a cross-sample mini-batch construction strategy that enables the student to perceive stable and generalized geometric structure. This aligns across diverse point cloud instances of the teacher, rather than within a single sample. Additionally, KL divergence is applied to align semantic distributions, and ground-truth supervision further reinforces accurate segmentation. Our method achieves state of the art performance with significantly reduced model complexity, demonstrating its effectiveness and efficiency in real-world deployment scenarios. Our Code is available at https://github.com/itsnotacie/SRKD.
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。