Research map / Detection and segmentation / CNN features
Dynamic Snake Convolution Based on Topological Geometric Constraints for Tubular Structure Segmentation
Yaolei Qi, Yuting He, Xiaoming Qi, Yuan Zhang, Guanyu Yang
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 |
|---|---|---|---|
| replace UNet with DSCNet (dynamic snake convolution) under cross-entropy (ablation) | UNet · DRIVE | Dice (%): 80.73 -> 81.85 | Improved |
| add TCLoss to UNet (ablation) | UNet · DRIVE | Betti error beta0: 1.209 -> 1.117 | Improved |
| add TCLoss to DSCNet (ablation) | DSCNet · DRIVE | Betti error beta0: 1.094 -> 0.998 | Improved |
| add TCLoss to DSCNet (ablation) | DSCNet · DRIVE | Betti error beta1: 0.780 -> 0.803 | Hurt |
| add TCLoss to DSCNet (ablation) | DSCNet · DRIVE | Hausdorff distance: 6.68 -> 6.78 | Hurt |
| add TCLoss to DSCNet (ablation) | DSCNet · ROADS | Hausdorff distance: 7.96 -> 7.34 | Improved |
| use proposed Hausdorff-based TCLoss instead of Wasserstein-based loss on UNet (ablation) | UNet · DRIVE | Betti error beta0: 1.144 -> 1.117 | Improved |
| add TCLoss to CS2-Net (ablation) | CS2-Net · DRIVE | Dice (%): 77.53 -> 79.69 | Improved |
| use DSCNet instead of UNet on 3D coronary data (ablation) | UNet · CORONARY (3D Cardiac CCTA) | Dice (%): 76.87 -> 79.92 | Improved |
| add TCLoss to DSCNet (ablation) | DSCNet · CORONARY (3D Cardiac CCTA) | Hausdorff distance: 6.326 -> 5.787 | Improved |
Closest papers by meaning
- HarmonySeg: Tubular Structure Segmentation with Deep-Shallow Feature Fusion and Growth-Suppression Balanced Loss (ICCV 2025 · similarity 0.90)
- GraphMorph: Tubular Structure Extraction by Morphing Predicted Graphs (NeurIPS 2024 · similarity 0.90)
- Segmentation of Tubular Structures Using Iterative Training with Tailored Samples (ICCV 2023 · similarity 0.88)
- TopoTTA: Topology-Enhanced Test-Time Adaptation for Tubular Structure Segmentation (ICCV 2025 · similarity 0.88)
- clDice - A Novel Topology-Preserving Loss Function for Tubular Structure Segmentation (CVPR 2021 · similarity 0.88)
- Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures (ECCV 2024 · similarity 0.88)
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.
