.png){kind=icon}
.png){kind=icon}
Return
GloW-VSNet: A Scribble-Based Weakly Supervised Framework for Global-View Vitiligo Lesion Segmentation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.media.2025.103920.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose GloW-VSNet, a scribble-guided weakly supervised framework for vitiligo segmentation. • Physician-provided scribbles guide spatial attention and enable effective lesion localization. • Differentiable feature clustering and spatial continuity optimization improve segmentation accuracy. • GloW-VSNet achieves state-of-the-art results across public and private global-view vitiligo datasets. • This is the first study to leverage scribble annotations for global-view vitiligo segmentation.
Keywords:
vitiligo segmentation
weakly supervised learning
spatial attention
scribble-based annotation
computer-aided diagnosis
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
11.8
Papers:
3.8K
Citations:
2.4W