Multinex introduces a multi-prior Retinex residual framework that produces lightweight (45K params) and nano (0.7K params) models for low-light image enhancement, outperforming other lightweight methods while approaching heavy-model performance.
Jobson, Zia-ur Rahman, and Glenn A
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
-
Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex
Multinex introduces a multi-prior Retinex residual framework that produces lightweight (45K params) and nano (0.7K params) models for low-light image enhancement, outperforming other lightweight methods while approaching heavy-model performance.