The author introduces an edge-guided Retinex model for enhancing low-light images using a novel inertial Bregman alternating linearized minimization algorithm.
Proposing LoLiSRFlow for joint low-light enhancement and super-resolution tasks using a transformer-based conditional flow network.
Proposing a zero-reference low-light enhancement framework using physical quadruple priors to achieve superior performance in various scenarios.
Utilizing pre-trained latent diffusion models to enhance neural ISP for low-light images.
Effiziente Bildverbesserung bei schwachem Licht durch die Kombination von Troublemaker-Lernen (TML) und Global Dynamic Convolution (GDC).
LYT-Net, a lightweight transformer-based network, achieves state-of-the-art performance on low-light image enhancement tasks while maintaining high computational efficiency.
A novel and robust low-light image enhancement method using CLIP-Fourier Guided Wavelet Diffusion (CFWD) that leverages multimodal visual-language information in the frequency domain to effectively bridge the gap between degraded and normal domains.
A novel trainable color space called Horizontal/Vertical-Intensity (HVI) is proposed to decouple image brightness and color, enabling efficient low-light image enhancement.
A novel Retinex-based Mamba architecture that leverages the computational efficiency of State Space Models and a Fused-Attention mechanism to effectively enhance low-light images while maintaining image quality.
The proposed CPGA-Net+ model achieves state-of-the-art performance in low-light image enhancement by incorporating an attention mechanism driven by the Atmospheric Scattering Model and integrating gamma correction into the local processing branch.