.png){kind=icon}
.png){kind=icon}
Return
RFD: A Reducing Feature Discrepancy method for unsupervised cross-modality SAM adaptation
.png){kind=icon}
PRE
AI
J
Z
W
Y
X
W
D
C
薛
S
R
DOI：10.1016/j.compmedimag.2026.102781.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Novel SAM-based framework Reducing Feature Discrepancy for cross-modality UDA. • Structural Prototype-based Contrastive Learning for robust domain connection. • Iterative Pixel-Prototype Transport for efficient online prototype clustering. • Reweighted Unbalanced Optimal Transport for precise feature alignment. • Superior performance on four challenging diverse cross-modality datasets.
Journal
IF:
4.9
Papers:
2.4K
Citations:
5.0K
Organization
Cited Papers