MAFI-Net: A Multi-level Attention Feature Interaction Network for Retinal Vessel Segmentation
MAFI-Net: A Multi-level Attention Feature Interaction Network for Retinal Vessel Segmentation
Xuelin Xu,Jianwei Chen,Bingfeng Ruan,Huabin He
2026 · DOI: 10.1007/s11760-026-05336-z
Signal, Image and Video Processing · 0 Citations
TLDR
A novel Multi-level Attention Feature Interaction Network (MAFI-Net) and a Multi-Constraint Guided Loss (MCGL) are proposed, which introduces constraints related to vessel skeleton and boundary information on top of traditional loss factors, thereby achieving more robust structural optimization.