When Extreme Darkness Meets Motion Blur: MeanFlow for Unified RAW Restoration

2026-08-03

AI Summary

This work addresses the common oversight in existing extremely low-light RAW image enhancement methods, which typically neglect the coupled degradation of motion blur and sensor noise present in real-world captures. To tackle this, the study introduces a unified framework that explicitly models realistic motion blur for the first time in this task, integrating a domain-conditioned RAW tokenizer with a single-step MeanFlow enhancement architecture to jointly perform deblurring and illumination enhancement. Furthermore, a physics-guided refinement module is devised, imposing zero additional inference cost while improving consistency between illumination and reflectance and enhancing color fidelity. Evaluated on the newly curated SIDED dataset—featuring controllable motion blur and authentic sensor noise—the proposed method significantly outperforms current approaches under complex degradation scenarios, achieving state-of-the-art performance.

Abstract

Extremely low-light RAW enhancement aims to recover severely attenuated sensor signals, yet existing methods often focus on illumination and noise while overlooking the motion-induced degradations inherent in practical low-light imaging. We present a framework for robust extremely low-light RAW enhancement under realistic acquisition degradations. First, we introduce See in the Degraded Extremely Dark (SIDED), a new dataset that applies controlled motion degradation to extremely low-light RAW pairs while retaining their original sensor noise. Second, we propose a unified RAW tokenizer equipped with explicit domain-conditioned representation calibration to align extremely low-light and well-exposed RAW data, followed by a MeanFlow that performs enhancement in a single function evaluation. To our knowledge, this is the first work to formulate extremely low-light RAW enhancement under realistic motion-degraded acquisition and address it with MeanFlow. We further introduce a physics-guided refinement model to strengthen illumination--reflectance consistency, pixel fidelity, and color preservation without incurring additional inference cost. Extensive experiments demonstrate that our framework achieves state-of-the-art performance in extremely low-light RAW enhancement, and robustly handles coupled motion and noise degradations.

Problem

• extremely low-light
• motion blur
• RAW enhancement
• image restoration
• realistic degradation

Innovation

• MeanFlow
• RAW restoration
• motion blur
• extremely low-light imaging
• domain-conditioned calibration