.png){kind=icon}
.png){kind=icon}
Return
CWE-Net: Cross weather expert network for adaptive image restoration
.png){kind=icon}
OA
AI
M
I
G
DOI：10.1016/j.imavis.2026.106068.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Modular expert-based framework for unified weather image restoration. • Multi-label router activates haze, rain, and snow experts. • Cross-weather fusion enables cooperative expert interaction. • Coordinate attention captures spatially structured weather artifacts. • Robust performance on synthetic and real-world benchmarks.
Keywords:
All-in-one image restoration
Adverse weather conditions
Mixture-of-experts
Cross-weather fusion
Coordinate attention
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.2
Papers:
4.0K
Citations:
6.7K
Organization
No organization information available
Cited Papers