LPD: Learnable Prototypes with Diversity Regularization for Weakly Supervised Histopathology Segmentation
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_ | 1866912751926902784 |
|---|---|
| author | Le, Khang Vu, Anh Mai Vo, Thi Kim Trang Thach, Ha Quang, Ngoc Bui Lam Nguyen, Thanh-Huy Le, Minh H. N. Han, Zhu Mohan, Chandra Van Nguyen, Hien |
| author_facet | Le, Khang Vu, Anh Mai Vo, Thi Kim Trang Thach, Ha Quang, Ngoc Bui Lam Nguyen, Thanh-Huy Le, Minh H. N. Han, Zhu Mohan, Chandra Van Nguyen, Hien |
| contents | Weakly supervised semantic segmentation (WSSS) in histopathology reduces pixel-level labeling by learning from image-level labels, but it is hindered by inter-class homogeneity, intra-class heterogeneity, and CAM-induced region shrinkage (global pooling-based class activation maps whose activations highlight only the most distinctive areas and miss nearby class regions). Recent works address these challenges by constructing a clustering prototype bank and then refining masks in a separate stage; however, such two-stage pipelines are costly, sensitive to hyperparameters, and decouple prototype discovery from segmentation learning, limiting their effectiveness and efficiency. We propose a cluster-free, one-stage learnable-prototype framework with diversity regularization to enhance morphological intra-class heterogeneity coverage. Our approach achieves state-of-the-art (SOTA) performance on BCSS-WSSS, outperforming prior methods in mIoU and mDice. Qualitative segmentation maps show sharper boundaries and fewer mislabels, and activation heatmaps further reveal that, compared with clustering-based prototypes, our learnable prototypes cover more diverse and complementary regions within each class, providing consistent qualitative evidence for their effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_05922 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LPD: Learnable Prototypes with Diversity Regularization for Weakly Supervised Histopathology Segmentation Le, Khang Vu, Anh Mai Vo, Thi Kim Trang Thach, Ha Quang, Ngoc Bui Lam Nguyen, Thanh-Huy Le, Minh H. N. Han, Zhu Mohan, Chandra Van Nguyen, Hien Computer Vision and Pattern Recognition Weakly supervised semantic segmentation (WSSS) in histopathology reduces pixel-level labeling by learning from image-level labels, but it is hindered by inter-class homogeneity, intra-class heterogeneity, and CAM-induced region shrinkage (global pooling-based class activation maps whose activations highlight only the most distinctive areas and miss nearby class regions). Recent works address these challenges by constructing a clustering prototype bank and then refining masks in a separate stage; however, such two-stage pipelines are costly, sensitive to hyperparameters, and decouple prototype discovery from segmentation learning, limiting their effectiveness and efficiency. We propose a cluster-free, one-stage learnable-prototype framework with diversity regularization to enhance morphological intra-class heterogeneity coverage. Our approach achieves state-of-the-art (SOTA) performance on BCSS-WSSS, outperforming prior methods in mIoU and mDice. Qualitative segmentation maps show sharper boundaries and fewer mislabels, and activation heatmaps further reveal that, compared with clustering-based prototypes, our learnable prototypes cover more diverse and complementary regions within each class, providing consistent qualitative evidence for their effectiveness. |
| title | LPD: Learnable Prototypes with Diversity Regularization for Weakly Supervised Histopathology Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.05922 |