Wakeup-Darkness: When Multimodal Meets Unsupervised Low-Light Image Enhancement

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Low-light image enhancement is a crucial visual task, and many unsupervised methods overlook the degradation of visible information in low-light scenes, adversely affecting the fusion of complementary information and hindering the generation of satisfactory results. To address this, we introduce Wakeup-Darkness, a multimodal enhancement framework that innovatively enriches user interaction through voice and textual commands. This approach signifies a technical leap and represents a paradigm shift in user engagement. We introduce a Cross-Modal Feature Fusion (CMFF) that synergizes semantic and depth context with low-light enhancement operations. Moreover, we propose a Gated Residual Block (GRB) and a channel-aware Look-Up Table (LUT) to adjust the intensity distribution of each channel. Crucially, the proposed Wakeup-Darkness scheme demonstrates remarkable generalization in unsupervised scenarios. The source code can be accessed from https://github.com/zhangbaijin/Wakeup-Dakness .
Funding Information
  • National Natural Science Foundation (62273235)
  • National Major Scientific Research Instrument Development (62227811)
  • Ministry of Education (8091B02210)

This publication has 32 references indexed in Scilit: