# Unprocessing Images for Learned Raw Denoising

arXiv: https://arxiv.org/abs/1811.11127 · HTML: https://mlsanity.com/paper/1811.11127/

## Key facts

- Cited by 250 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Nov 2018.

## Built on by (most cited first, 30 of 250)

- [Multi-Stage Progressive Image Restoration](https://mlsanity.com/paper/2102.02808/) (2021)
- [Designing a Practical Degradation Model for Deep Blind Image Super-Resolution](https://mlsanity.com/paper/2103.14006/) (2021)
- [NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Images](https://mlsanity.com/paper/2111.13679/) (2021)
- [Learning Enriched Features for Fast Image Restoration and Enhancement](https://mlsanity.com/paper/2205.01649/) (2022)
- [Deep Generalized Unfolding Networks for Image Restoration](https://mlsanity.com/paper/2204.13348/) (2022)
- [Invertible Image Signal Processing](https://mlsanity.com/paper/2103.15061/) (2021)
- [Practical Blind Image Denoising via Swin-Conv-UNet and Data Synthesis](https://mlsanity.com/paper/2203.13278/) (2022)
- [Deep Burst Super-Resolution](https://mlsanity.com/paper/2101.10997/) (2021)
- [You Only Need 90K Parameters to Adapt Light: A Light Weight Transformer for Image Enhancement and Exposure Correction](https://mlsanity.com/paper/2205.14871/) (2022)
- [Image Denoising: The Deep Learning Revolution and Beyond -- A Survey Paper --](https://mlsanity.com/paper/2301.03362/) (2023)
- [Invertible Denoising Network: A Light Solution for Real Noise Removal](https://mlsanity.com/paper/2104.10546/) (2021)
- [Physics-based Noise Modeling for Extreme Low-light Photography](https://mlsanity.com/paper/2108.02158/) (2021)
- [Low-light Image Enhancement via Breaking Down the Darkness](https://mlsanity.com/paper/2111.15557/) (2021)
- [Burst Image Restoration and Enhancement](https://mlsanity.com/paper/2110.03680/) (2021)
- [Rethinking Noise Synthesis and Modeling in Raw Denoising](https://mlsanity.com/paper/2110.04756/) (2021)
- [Zero-Shot Noise2Noise: Efficient Image Denoising without any Data](https://mlsanity.com/paper/2303.11253/) (2023)
- [Multitask AET with Orthogonal Tangent Regularity for Dark Object Detection](https://mlsanity.com/paper/2205.03346/) (2022)
- [Masked Image Training for Generalizable Deep Image Denoising](https://mlsanity.com/paper/2303.13132/) (2023)
- [Realistic Blur Synthesis for Learning Image Deblurring](https://mlsanity.com/paper/2202.08771/) (2022)
- [Dynamic Attentive Graph Learning for Image Restoration](https://mlsanity.com/paper/2109.06620/) (2021)
- [Dancing under the stars: video denoising in starlight](https://mlsanity.com/paper/2204.04210/) (2022)
- [Instance Segmentation in the Dark](https://mlsanity.com/paper/2304.14298/) (2023)
- [Model-Based Image Signal Processors via Learnable Dictionaries](https://mlsanity.com/paper/2201.03210/) (2022)
- [Deep Reparametrization of Multi-Frame Super-Resolution and Denoising](https://mlsanity.com/paper/2108.08286/) (2021)
- [IDR: Self-Supervised Image Denoising via Iterative Data Refinement](https://mlsanity.com/paper/2111.14358/) (2021)
- [RobustSAM: Segment Anything Robustly on Degraded Images](https://mlsanity.com/paper/2406.09627/) (2024)
- [Aleth-NeRF: Illumination Adaptive NeRF with Concealing Field Assumption](https://mlsanity.com/paper/2312.09093/) (2023)
- [Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving](https://mlsanity.com/paper/2404.04804/) (2024)
- [JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image Restoration](https://mlsanity.com/paper/2504.04158/) (2025)
- [BSRT: Improving Burst Super-Resolution with Swin Transformer and Flow-Guided Deformable Alignment](https://mlsanity.com/paper/2204.08332/) (2022)

## Abstract (arXiv)

Machine learning techniques work best when the data used for training resembles the data used for evaluation. This holds true for learned single-image denoising algorithms, which are applied to real raw camera sensor readings but, due to practical constraints, are often trained on synthetic image data. Though it is understood that generalizing from synthetic to real data requires careful consideration of the noise properties of image sensors, the other aspects of a camera's image processing pipeline (gain, color correction, tone mapping, etc) are often overlooked, despite their significant effect on how raw measurements are transformed into finished images. To address this, we present a technique to "unprocess" images by inverting each step of an image processing pipeline, thereby allowing us to synthesize realistic raw sensor measurements from commonly available internet photos. We additionally model the relevant components of an image processing pipeline when evaluating our loss function, which allows training to be aware of all relevant photometric processing that will occur after denoising. By processing and unprocessing model outputs and training data in this way, we are able to train a simple convolutional neural network that has 14%-38% lower error rates and is 9x-18x faster than the previous state of the art on the Darmstadt Noise Dataset, and generalizes to sensors outside of that dataset as well.