Under Construction
Retinal vessel segmentation is critical for medical diagnosis, yet existing models often struggle to generalize across domains due to appearance variability, limited annotations, and complex vascular morphology. We propose GraphSeg, a variational Bayesian framework that integrates anatomical graph priors with structure-aware image decomposition to enhance cross-domain segmentation. GraphSeg factorizes retinal images into structure-preserved and structure-degraded components, enabling domain-invariant representation. A deformable graph prior, derived from a statistical retinal atlas, is incorporated via a differentiable alignment and guided by an unsupervised energy function. Experiments on three public benchmarks (CHASE, DRIVE, HRF) show that GraphSeg consistently outperforms existing methods under domain shifts. These results highlight the importance of jointly modeling anatomical topology and image structure for robust generalizable vessel segmentation.
- ✅ Release the main code
- Reconstruct the code structure
- Add Instructions for Data Preparation
- Add Instructions for Training and Evaluation
- Add Pretrained Models
- Add Results and Visualizations
If you find this work useful in your research, please consider citing:
@inproceedings{liu2025towards,
title={Towards Generalizable Retina Vessel Segmentation with Deformable Graph Priors},
author={Ke Liu and Shangde Gao and Yichao Fu and Shangqi Gao},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=zVkbsGlKn9}
}We use the code from the following repositories for graph construction:
We appreciate the authors for sharing their code.