Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910964338655232 |
|---|---|
| author | Ma, Shuxian Dong, Zihao Cong, Runmin Kwong, Sam Shao, Xiuli |
| author_facet | Ma, Shuxian Dong, Zihao Cong, Runmin Kwong, Sam Shao, Xiuli |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17666 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification Ma, Shuxian Dong, Zihao Cong, Runmin Kwong, Sam Shao, Xiuli Computer Vision and Pattern Recognition I.4.0; I.5.0 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. |
| title | Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification |
| topic | Computer Vision and Pattern Recognition I.4.0; I.5.0 |
| url | https://arxiv.org/abs/2505.17666 |