.png){kind=icon}
.png){kind=icon}
Return
Distinct Polyp Generator Network for polyp segmentation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.imavis.2025.105847.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A feature progressive enhancement module is designed to address the problem of fuzzy polyp boundaries. • A dynamical aggregation module is proposed to overcome the influence of polyp size variations on segmentation. • A distinct polyp generator network integrating the above two modules is proposed for enhancing polyp segmentation performance, particularly for polyps with blurred boundaries and various sizes. • Extensive experiments over five polyp image datasets demonstrate the superiority of the proposed model, and its cross-domain generalization ability is verified on the related skin lesion segmentation task.
Journal
IF:
4.2
Papers:
4.0K
Citations:
6.7K