Research map / Detection and segmentation / CNN features
Joint Topology-Preserving and Feature-Refinement Network for Curvilinear Structure Segmentation
Mingfei Cheng, Kaili Zhao, Xuhong Guo, Yajing Xu, Jun Guo
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 |
|---|---|---|---|
| add boundary detection decoder (ablation) | Base UNet · CrackTree200 | F1: 78.86 -> 80.27 | Improved |
| add boundary detection decoder (ablation) | Base UNet · CrackTree200 | Quality: 77.56 -> 78.79 | Improved |
| add Feature Interactive Module to boundary branch (ablation) | Base-BO · CrackTree200 | F1: 80.27 -> 82.07 | Improved |
| add Feature Interactive Module to boundary branch (ablation) | Base-BO · CrackTree200 | Quality: 78.79 -> 80.75 | Improved |
| add Gated Attentive Unit in skip connections (ablation) | Base UNet · CrackTree200 | F1: 78.86 -> 81.84 | Improved |
| add Gated Attentive Unit in skip connections (ablation) | Base UNet · CrackTree200 | Quality: 77.56 -> 81.55 | Improved |
| replace concatenation with GAU (ablation) | Base-C · CrackTree200 | F1: 80.44 -> 81.84 | Improved |
| combine FIM and GAU with iterative feedback (ablation) | Base-FIM · CrackTree200 | F1: 82.07 -> 84.19 | Improved |
| combine FIM and GAU with iterative feedback (ablation) | Base-GAU · CrackTree200 | F1: 81.84 -> 84.19 | Improved |
| use 3 feedback iterations instead of 1 (ablation) | JTFN · CrackTree200 | F1: 82.30 -> 84.19 | Improved |
| use 4 feedback iterations instead of 3 (ablation) | JTFN · CrackTree200 | F1 | Hurt |
| JTFN versus DRU on Crack500 recall | JTFN · Crack500 | Recall: 71.43 -> 69.06 | Hurt |
Closest papers by meaning
- Joint Segmentation and Path Classification of Curvilinear Structures (TPAMI 2019 · similarity 0.89)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.89)
- From Reconstruction to Decision: A Post-Encoder Plug-in Adapter for Curvilinear Segmentation (ECCV 2026 · similarity 0.87)
- Beyond the Pixel-Wise Loss for Topology-Aware Delineation (CVPR 2018 · similarity 0.87)
- SegGraph: Leveraging Graphs of SAM Segments for Few-Shot 3D Part Segmentation (NeurIPS 2025 · similarity 0.87)
- DeepBranchTracer: A Generally-Applicable Approach to Curvilinear Structure Reconstruction Using Multi-Feature Learning (AAAI 2024 · similarity 0.86)
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.
