Multinex: Lightweight Low-light Image Enhancement via Multi-prior Retinex
For edge deployment of low-light enhancement, Multinex provides a practical solution that drastically reduces model size while maintaining high performance.
Multinex proposes an ultra-lightweight framework for low-light image enhancement that integrates multiple color space priors within a Retinex formulation, achieving SOTA performance with only 45K or even 0.7K parameters, significantly outperforming other lightweight models and matching heavy models.
Low-light image enhancement (LLIE) aims to restore natural visibility, color fidelity, and structural detail under severe illumination degradation. State-of-the-art (SOTA) LLIE techniques often rely on large models and multi-stage training, limiting practicality for edge deployment. Moreover, their dependence on a single color space introduces instability and visible exposure or color artifacts. To address these, we propose Multinex, an ultra-lightweight structured framework that integrates multiple fine-grained representations within a principled Retinex residual formulation. It decomposes an image into illumination and color prior stacks derived from distinct analytic representations, and learns to fuse these representations into luminance and reflectance adjustments required to correct exposure. By prioritizing enhancement over reconstruction and exploiting lightweight neural operations, Multinex significantly reduces computational cost, exemplified by its lightweight (45K parameters) and nano (0.7K parameters) versions. Extensive benchmarks show that all lightweight variants significantly outperform their corresponding lightweight SOTA models, and reach comparable performance to heavy models. Paper page available at https://albrateanu.github.io/multinex.