Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Model

TL;DR

The paper introduces LED, a novel RAW image denoising pipeline that implicitly calibrates the denoiser instead of the noise model, enabling fast, adaptable, and effective denoising across various camera models and lighting conditions.

Contribution

It proposes an implicit fine-tuning approach for denoisers that eliminates explicit calibration, reducing labor, enhancing transferability, and improving performance in low-light RAW denoising.

Findings

LED outperforms existing methods across multiple camera models.

Requires only minimal data and iterations for effective denoising.

Reduces synthetic-real noise disparity without extra computational cost.

Abstract

Explicit calibration-based methods have dominated RAW image denoising under extremely low-light environments. However, these methods are impeded by several critical limitations: a) the explicit calibration process is both labor- and time-intensive, b) challenge exists in transferring denoisers across different camera models, and c) the disparity between synthetic and real noise is exacerbated by digital gain. To address these issues, we introduce a groundbreaking pipeline named Lighting Every Darkness (LED), which is effective regardless of the digital gain or the camera sensor. LED eliminates the need for explicit noise model calibration, instead utilizing an implicit fine-tuning process that allows quick deployment and requires minimal data. Structural modifications are also included to reduce the discrepancy between synthetic and real noise without extra computational demands. Our method surpasses existing methods in various camera models, including new ones not in public datasets, with just a few pairs per digital gain and only 0.5% of the typical iterations. Furthermore, LED also allows researchers to focus more on deep learning advancements while still utilizing sensor engineering benefits. Code and related materials can be found in https://srameo.github.io/projects/led-iccv23/ .