What question did this study set out to answer?

This research aims to enhance low-light images while avoiding common artifacts and preserving details.

Overlapped Wavelet Diffusion for Low-Light Image Enhancement

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates improved image quality in low-light conditions, suggesting enhanced visual performance.

This research aims to enhance low-light images while avoiding common artifacts and preserving details.
Proposed an overlapped wavelet diffusion framework for LLIE using OWT and HFEBlock.
Conducted extensive experiments on the LOLv1 and LOLv2-real datasets to evaluate performance and artifacts.
Compared the proposed OWDiff framework against existing LLIE methods in terms of visual quality and computational efficiency.
Achieved an average PSNR gain of 0.58 dB compared to DiffLL.
Obtained a 1.64% relative improvement in SSIM on both datasets.
Demonstrated a 5.9% relative reduction in LPIPS, indicating better texture quality.

彭芬 et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark