A Transformer architecture with a learned dynamic integral-image illumination estimator and illumination-guided attention achieves small but consistent PSNR and SSIM gains over prior low-light enhancement methods on six public datasets.
Naturalness preserved enhancement algorithm for non-uniform illumination images,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SAIGFormer: A Spatially-Adaptive Illumination-Guided Network for Low-Light Image Enhancement
A Transformer architecture with a learned dynamic integral-image illumination estimator and illumination-guided attention achieves small but consistent PSNR and SSIM gains over prior low-light enhancement methods on six public datasets.