Research map / Detection and segmentation / Weakly supervised detection and segmentation
Bridging RGB and Hematoxylin Components: An Interleaved Guidance and Fusion Framework for Point Supervised Nuclei Segmentation
Zihan Huan, Xipeng Pan, Hualong Zhang, Siyang Feng, Rushi Lan, Huadeng Wang, Haoxiang Lu, Zhenbing Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fuse RGB and Hematoxylin inputs with ECAU instead of a single input (ablation) | DFGNet baseline · MoNuSeg (MO) | Acc (%): 90.13 -> 90.34 | Improved |
| use Hematoxylin input instead of RGB input alone (ablation) | single-input baseline · MoNuSeg (MO) | Acc (%): 90.13 -> 90.05 | No clear effect |
| add all modules (RCDF, IGA, interleaved learning, ECAU) to dual input (ablation) | DFGNet · MoNuSeg (MO) | Dice_obj (%): 73.24 -> 74.89 | Improved |
| insert IGA in more decoder layers (1 to 4) (ablation) | DFGNet with RCDF+ECAU · MoNuSeg (MO) | AJI: 53.17 -> 54.85 | Improved |
| fuse dual branches instead of using the RGB branch alone | DFGNet · CPM17 | PQ: 50.27 -> 52.16 | Improved |
| dual-representation DFGNet versus best prior point-supervised method PUCD | DFGNet · MoNuSeg (MO) | AJI: 50.43 -> 54.85 | Improved |
| DFGNet on CoNSeP versus baselines | DFGNet · CoNSeP | Acc: 90.13 (SCNet) vs 89.88 | Hurt |
| vary loss weights lambda1 and lambda2 (ablation) | DFGNet · MoNuSeg (MO) | F1-score, Dice_obj: within 3% | No clear effect |
| perturb point labels up to 20 pixels (ablation) | DFGNet · point shift | Acc, F1, Dice, AJI | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SCNet — SCNet: Training Inference Sample Consistency for Instance Segmentation (AAAI 2021) · better on 1 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning to Generalize over Subpartitions for Heterogeneity-Aware Domain Adaptive Nuclei Segmentation (IJCV 2024 · similarity 0.87)
- Cross-Patch Dense Contrastive Learning for Semi-Supervised Segmentation of Cellular Nuclei in Histopathologic Images (CVPR 2022 · similarity 0.86)
- Sparse Multi-Modal Graph Transformer With Shared-Context Processing for Representation Learning of Giga-Pixel Images (CVPR 2023 · similarity 0.86)
- SheafStain: Sheaf-Theoretic Schr\"odinger Bridge for Spatially and Biologically Coherent Virtual Staining (NeurIPS 2026 · similarity 0.85)
- Semantic and Visual Crop-Guided Diffusion Models for Heterogeneous Tissue Synthesis in Histopathology (NeurIPS 2025 · similarity 0.84)
- Dynamic Entity-Masked Graph Diffusion Model for Histopathology Image Representation Learning (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
