arXiv 2505.176662025-05-23cs.CV

Proto-FG3D：基于原型可解释的细粒度三维形状分类

Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification

School of Information Science and Engineering（信息科学与工程学院）；University of Jinan（济南大学）；School of Control Science and Engineering（控制科学与工程学院）；Shandong University（山东大学）；Department of Computing and Decision Sciences（计算与决策科学系）；Lingnan University（岭南大学）；College of Computer Science（计算机科学学院）；Nankai University（南开大学）

Shuxian Ma, Zihao Dong, Runmin Cong, Sam Kwong, Xiuli Shao

备注：: 11 pages, 2 figures, 5 tablets; Submitted to BMVC2025

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对细粒度三维形状分类中判别信息有限、类别不平衡及可解释性不足的问题，提出首个基于原型的框架Proto-FG3D，通过原型关联、在线聚类和原型引导监督学习实现非参数化原型学习，在FG3D和ModelNet40上超越现有方法。

摘要：

基于深度学习的多视图粗粒度三维形状分类在过去十年中取得了显著成功，这得益于CNN和ViT骨干网络强大的特征学习能力。然而，作为对详细形状理解至关重要的一个具有挑战性的研究领域，细粒度三维分类仍研究不足，原因在于多视图特征聚合过程中捕获的判别信息有限，特别是对于细微的类间差异、类别不平衡以及参数化模型固有的可解释性限制。为解决这些问题，我们提出了首个基于原型的框架Proto-FG3D，用于细粒度三维形状分类，实现了从参数化softmax到非参数化原型学习的范式转变。首先，Proto-FG3D通过原型关联建立了联合多视图和多类别表示学习。其次，通过在线聚类对原型进行细化，提高了多视图特征分配的鲁棒性和子类间平衡。最后，建立了原型引导的监督学习，通过原型-视图相关性分析增强细粒度判别能力，并通过透明的基于案例的推理实现即席可解释性。在FG3D和ModelNet40上的实验表明，Proto-FG3D在准确性、透明预测和带可视化的即席可解释性方面超越了最先进的方法，挑战了传统的细粒度三维识别方法。

英文摘要：

Deep learning-based multi-view coarse-grained 3D shape classification has achieved remarkable success over the past decade, leveraging the powerful feature learning capabilities of CNN-based and ViT-based backbones. However, as a challenging research area critical for detailed shape understanding, fine-grained 3D classification remains understudied due to the limited discriminative information captured during multi-view feature aggregation, particularly for subtle inter-class variations, class imbalance, and inherent interpretability limitations of parametric model. To address these problems, we propose the first prototype-based framework named Proto-FG3D for fine-grained 3D shape classification, achieving a paradigm shift from parametric softmax to non-parametric prototype learning. Firstly, Proto-FG3D establishes joint multi-view and multi-category representation learning via Prototype Association. Secondly, prototypes are refined via Online Clustering, improving both the robustness of multi-view feature allocation and inter-subclass balance. Finally, prototype-guided supervised learning is established to enhance fine-grained discrimination via prototype-view correlation analysis and enables ad-hoc interpretability through transparent case-based reasoning. Experiments on FG3D and ModelNet40 show Proto-FG3D surpasses state-of-the-art methods in accuracy, transparent predictions, and ad-hoc interpretability with visualizations, challenging conventional fine-grained 3D recognition approaches.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。