Search papers, labs, and topics across Lattice.
This paper introduces the Conditional Mutual Information-Guided Network (CMIG-Net) for low-light image enhancement, addressing the limitations of existing dual-branch architectures like CIDNet that fail to account for the mutual interaction between intensity and chrominance. By employing a Conditional Mutual Information Calibration (CMIC) module, CMIG-Net generates a conditional information map that allows for adaptive recalibration of chrominance based on local illumination conditions. Experimental results show that CMIG-Net outperforms CIDNet, achieving significant improvements in PSNR, particularly on challenging datasets like Sony-Total-Dark.
CMIG-Net achieves up to a 0.619 dB gain in PSNR over existing methods by effectively leveraging conditional mutual information for low-light image enhancement.
Low-light image enhancement (LLIE) seeks to restore structural fidelity, natural color rendition, and proper exposure from images captured under inadequate lighting conditions. Recent state-of-the-art approaches, such as CIDNet, adopt a dual-branch architecture comprising a chrominance (HV) branch and an intensity (I) branch to separately model decoupled chromatic and luminance information within the HVI color space. However, these methods overlook the mutual interaction between intensity and chrominance components, which inherently limits their representational capacity and leads to suboptimal enhancement performance. To address this limitation, we propose the Conditional Mutual Information-Guided Network (CMIG-Net), which leverages conditional mutual information as a principled metric to quantitatively assess the contribution of chrominance features conditioned on the available intensity information. In particular, we design a Conditional Mutual Information Calibration (CMIC) module that generates a conditional information map, enabling region-adaptive recalibration of chrominance representations according to local illumination statistics. Furthermore, we introduce a Dynamic Dual-branch Information Restoration (D2IR) module, which adaptively governs bidirectional information flow between the intensity and chrominance branches, guided by both the conditional prior and the instantaneous restoration state. Extensive experiments on paired LLIE benchmarks demonstrate that CMIG-Net consistently outperforms CIDNet, achieving up to a 0.619 dB gain in PSNR, with a 0.382 dB improvement specifically on the challenging Sony-Total-Dark dataset.
Tsinghua AI