LEIQ-Assessor applies multi-task learning on a SigLIP2 backbone to jointly predict MOS and six attributes for low-light enhanced images using PLCC loss, outperforming prior IQA models on the MLE benchmark.
Dual-branch network for portrait image quality assessment,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LEIQ-Assessor: Multi-dimensional Quality Assessment of Low-light Enhanced Images via Multi-task Learning
LEIQ-Assessor applies multi-task learning on a SigLIP2 backbone to jointly predict MOS and six attributes for low-light enhanced images using PLCC loss, outperforming prior IQA models on the MLE benchmark.