HPGN is a hybrid priors-guided network that jointly enhances illumination and removes compression artifacts in low-light images by using JPEG QF and QM to guide plug-and-play modules, with a random QF training strategy for varying compression levels.
Datasets and Experimental Settings We use LOLv1 [21], LOLv2-real [22], and LOLv2-syn [22] as our benchmark datasets
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HPGN: Hybrid Priors-Guided Network for Compressed Low-Light Image Enhancement
HPGN is a hybrid priors-guided network that jointly enhances illumination and removes compression artifacts in low-light images by using JPEG QF and QM to guide plug-and-play modules, with a random QF training strategy for varying compression levels.