Scholarlink

<p>Low-light image enhancement remains a challenging task due to the loss of detail and color distortion under insufficient illumination. This paper introduces VMamba-LLIE, a novel triple-branch network that leverages Signal-to-Noise-Ratio prior guidance and Horizontal/Vertical Intensity color space assistance for superior low-light image enhancement. The proposed model integrates a global modeling branch for long-range dependency analysis, a local modeling branch for fine-grained detail recovery, and a color assistance branch for high-quality color reconstruction. Extensive experiments across diverse datasets demonstrate that VMamba-LLIE achieves state-of-the-art performance with a smaller model size, outperforming existing methods in both efficiency and effectiveness.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

VMamba-LLIE: enhancing low-light images with snr prior-guided and HVI color-assisted triple-branch network

  • 期刊论文

  • 发表日期：

    2026年7月16日

查看全文
  • Zhanqiang Huo,
  • Pengyun Shi,
  • Yizhang Meng,
  • Yingxu Qiao,
  • Shan Zhao

摘要

Low-light image enhancement remains a challenging task due to the loss of detail and color distortion under insufficient illumination. This paper introduces VMamba-LLIE, a novel triple-branch network that leverages Signal-to-Noise-Ratio prior guidance and Horizontal/Vertical Intensity color space assistance for superior low-light image enhancement. The proposed model integrates a global modeling branch for long-range dependency analysis, a local modeling branch for fine-grained detail recovery, and a color assistance branch for high-quality color reconstruction. Extensive experiments across diverse datasets demonstrate that VMamba-LLIE achieves state-of-the-art performance with a smaller model size, outperforming existing methods in both efficiency and effectiveness.