Research map / Detection and segmentation / CNN features
Region Focus Network for Joint Optic Disc and Cup Segmentation
Ge Li, Changsheng Li, Chan Zeng, Peng Gao, Guotong Xie
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 Pyramid Feature Pooling with Fusion Feature Pooling (ablation) | RF-Net · REFUGE val | DiceOC (%): 87.83 -> 89.79 | Improved |
| replace Pyramid Feature Pooling with Fusion Feature Pooling (ablation) | RF-Net · REFUGE val | mIoU (%): 85.76 -> 87.08 | Improved |
| use Focal+Dice loss instead of BCE loss (ablation) | RF-Net · REFUGE val | DiceOD (%): 93.84 -> 96.06 | Improved |
| add focal loss (weight 0.3) to Dice loss (ablation) | RF-Net · REFUGE val | mIoU (%): 84.73 -> 87.08 | Improved |
| add BCE loss to Dice loss (ablation) | RF-Net · REFUGE val | mIoU (%): 84.73 -> 84.10 | Hurt |
| use focal loss alone instead of Dice loss alone (ablation) | RF-Net · REFUGE val | mIoU (%): 84.73 -> 82.26 | Hurt |
| increase mask shape from 28x28 to 56x56 (ablation) | RF-Net · REFUGE val | DiceOD (%): 93.03 -> 96.06 | Improved |
| use unified detection plus multi-class mask branch instead of Mask R-CNN | optic disc/cup segmentation · REFUGE val | DiceOC (%): 82.36 -> 89.79 | Improved |
Closest papers by meaning
- SIGraph: Saliency Image-Graph Network for Retinal Disease Classification in Fundus Image (AAAI 2025 · similarity 0.82)
- HACDR-Net: Heterogeneous-Aware Convolutional Network for Diabetic Retinopathy Multi-Lesion Segmentation (AAAI 2024 · similarity 0.81)
- FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs (NeurIPS 2026 · similarity 0.81)
- Like an Ophthalmologist: Dynamic Selection Driven Multi-View Learning for Diabetic Retinopathy Grading (AAAI 2025 · similarity 0.80)
- Upping the Game: How 2D U-Net Skip Connections Flip 3D Segmentation (NeurIPS 2024 · similarity 0.80)
- Polyper: Boundary Sensitive Polyp Segmentation (AAAI 2024 · similarity 0.79)
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.
