Why Low-Light Cameras Go Color Blind: Removing Color Bias in Raw Denoising

2026-07-13

AI Summary

This work addresses the severe color distortion commonly observed in low-light raw image denoising, which is primarily caused by black level errors—a factor previously overlooked in the literature. The study is the first to identify black level error as the dominant source of color shifts in such scenarios and introduces a blind denoising method that requires neither camera calibration nor device-specific parameters, rendering it universally applicable. The proposed approach employs a global bias estimation network to correct black level errors and establishes an unbiased ground-truth extraction framework to rectify color biases in the SIDD dataset. Without paired training data or camera metadata, the deep network achieves high-fidelity color-preserving denoising, significantly outperforming existing blind methods on the ELD, SID, and LRID benchmarks, with certain metrics even matching or surpassing those of strongly supervised approaches.

Abstract

Raw images inherently suffer from noise due to the stochastic nature of light and sensor hardware imperfections. As real photon counts fall, the ratio of this noise to the signal degrades; consequently, for low-light conditions, robust denoising is especially vital for high-quality results. While recent data-driven methods achieve strong performance, they typically rely on large-scale noisy-clean image pairs that are costly and difficult to collect. Alternatively, parametric noise models can generate synthetic training data, but this necessitates precise camera calibration, which is often impractical for unknown devices. In this work, we propose a camera-agnostic, calibration-free paradigm for low-light raw denoising. We identify that color bias from black-level error is a primary source of performance degradation and causes severe color shifts. To mitigate this, we introduce a bias estimator network that predicts the black-level error as a global feature of the noisy input. We evaluate our approach across the ELD, SID, and LRID datasets, demonstrating superior performance among blind denoisers, particularly in terms of color correction. In many cases, we are competitive with-or can even surpass-methods with stronger supervision. Furthermore, we reveal that the widely used SIDD dataset contains significant color bias in its ground-truth images, which yields unrealistic color reproduction in trained models. We introduce a new ground-truth extraction framework to resolve this issue and provide a benchmark of existing methods on the corrected dataset.

Problem

• low-light denoising
• color bias
• black-level error
• raw image
• camera-agnostic

Innovation

• black-level error
• color bias
• camera-agnostic denoising
• raw image denoising
• ground-truth correction