Self-DACE++: Robust Low-Light Enhancement via Efficient Adaptive Curve Estimation

TL;DR

Self-DACE++ is a lightweight, unsupervised framework for low-light image enhancement that improves quality and efficiency through adaptive curves, a novel training strategy, and a physics-based loss, achieving real-time performance.

Contribution

It introduces enhanced adaptive adjustment curves and a randomized training strategy to improve low-light image enhancement efficiency and quality.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Achieves real-time inference with superior enhancement quality.

Effectively suppresses noise in dark regions using a dedicated denoising module.

Abstract

In this paper, we present Self-DACE++, an improved unsupervised and lightweight framework for Low-Light Image Enhancement (LLIE), building upon our previous Self-Reference Deep Adaptive Curve Estimation (Self-DACE). To better address the trade-off between computational efficiency and restoration quality, Self-DACE++ introduces enhanced Adaptive Adjustment Curves (AACs). These curves, governed by minimal trainable parameters, flexibly adjust the dynamic range while preserving the color fidelity, structural integrity, and naturalness of the enhanced images. To achieve an extremely lightweight architecture without sacrificing performance, we propose a randomized order training strategy coupled with a network fusion mechanism, which compresses the model into an efficient iterative inference structure. Furthermore, we formulate a physics-grounded objective function based on Retinex theory and incorporate a dedicated denoising module to effectively estimate and suppress latent noise in dark regions. Extensive qualitative and quantitative evaluations on multiple real-world benchmark datasets demonstrate that Self-DACE++ outperforms existing state-of-the-art methods, delivering superior enhancement quality with real-time inference capability. The code is available at https://github.com/John-Wendell/Self-DACE.