.png){kind=icon}
.png){kind=icon}
Return
BWNet: A bridged W-shaped network for hierarchical feature interaction in infrared small target detection
.png){kind=icon}
PRE
AI
M
J
B
Q
DOI：10.1016/j.neucom.2026.134712.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Propose BWNet with symmetric U-Net for continuous feature transfer and semantic enhancement. • Design a SUFEM enabling encoder-decoder symmetric interactions for multi-level feature collaboration. • Design an AMLFM to adaptively balance local details with global semantics using multi-level weights. • Achieves optimal balance: keeps performance high while minimizing model complexity and computation.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W
Organization
Cited Papers