Image demosaicking and denoising are the two key fundamental steps in digital
camera pipelines, aiming to reconstruct clean color images from noisy luminance
readings. In this paper, we propose and study Wild-JDD, a novel learning
framework for joint demosaicking and denoising in the wild. In contrast to
previous works which generally assume the ground truth of training data is a
perfect reflection of the reality, we consider here the more common imperfect
case of ground truth uncertainty in the wild. We first illustrate its
manifestation as various kinds of artifacts including zipper effect, color
moire and residual noise. Then we formulate a two-stage data degradation
process to capture such ground truth uncertainty, where a conjugate prior
distribution is imposed upon a base distribution. After that, we derive an
evidence lower bound (ELBO) loss to train a neural network that approximates
the parameters of the conjugate prior distribution conditioned on the degraded
input. Finally, to further enhance the performance for out-of-distribution
input, we design a simple but effective fine-tuning strategy by taking the
input as a weakly informative prior. Taking into account ground truth
uncertainty, Wild-JDD enjoys good interpretability during optimization.
Extensive experiments validate that it outperforms state-of-the-art schemes on
joint demosaicking and denoising tasks on both synthetic and realistic raw
datasets.

本研究提出和研究了一种名为 Wild-JDD 的新型联合去马赛克和去噪声的学习框架，并成功实现了对于数字相机管道中图像去马赛克和去噪声的高效处理。

野外联合去马赛克和降噪：在真实情况下降低地面真实性不确定性的情况

Joint Demosaicking and Denoising in the Wild: The Case of Training Under  Ground Truth Uncertainty

Image demosaicking and denoising are the two key steps for color image
production pipeline. The classical processing sequence consists of applying
denoising first, and then demosaicking. However, this sequence leads to
oversmoothing and unpleasant checkerboard effect. Moreover, it is very
difficult to change this order, because once the image is demosaicked, the
statistical properties of the noise will be changed dramatically. This is
extremely challenging for traditional denoising models that strongly rely on
statistical assumptions. In this paper, we attempt to tackle this prickly
problem. Indeed, here we invert the traditional CFA processing pipeline by
first demosaicking and then denoising. In the first stage, we design a
demosaicking algorithm that combines traditional methods and a convolutional
neural network (CNN) to reconstruct a full color image ignoring the noise. To
improve the performance in image demosaicking, we modify an Inception
architecture for fusing R, G and B three channels information. This stage
retains all known information that is the key point to obtain pleasurable final
results. After demosaicking, we get a noisy full-color image and use another
CNN to learn the demosaicking residual noise (including artifacts) of it, that
allows to obtain a restored full color image. Our proposed algorithm completely
avoids the checkerboard effect and retains more image detail. Furthermore, it
can process very high-level noise while the performances of other CNN based
methods for noise higher than 20 are rather limited. Experimental results show
clearly that our method outperforms state-of-the-art methods both
quantitatively as well as in terms of visual quality.

本文提出了一种优于当前最先进方法的图像去马赛克和降噪算法，该算法将传统的 CFA 处理流程中的处理顺序颠倒，首先进行图像去马赛克，然后进行降噪，避免了传统方法中的众多问题。算法利用卷积神经网络进行实现并具有处理高噪声的能力。

用于有效联合去马赛克 - 去噪的残差学习

Residual Learning for Effective joint Demosaicing-Denoising

While variational methods have been among the most powerful tools for solving
linear inverse problems in imaging, deep (convolutional) neural networks have
recently taken the lead in many challenging benchmarks. A remaining drawback of
deep learning approaches is their requirement for an expensive retraining
whenever the specific problem, the noise level, noise type, or desired measure
of fidelity changes. On the contrary, variational methods have a plug-and-play
nature as they usually consist of separate data fidelity and regularization
terms.
In this paper we study the possibility of replacing the proximal operator of
the regularization used in many convex energy minimization algorithms by a
denoising neural network. The latter therefore serves as an implicit natural
image prior, while the data term can still be chosen independently. Using a
fixed denoising neural network in exemplary problems of image deconvolution
with different blur kernels and image demosaicking, we obtain state-of-the-art
reconstruction results. These indicate the high generalizability of our
approach and a reduction of the need for problem-specific training.
Additionally, we discuss novel results on the analysis of possible optimization
algorithms to incorporate the network into, as well as the choices of algorithm
parameters and their relation to the noise level the neural network is trained
on.

本文研究探讨通过使用去噪神经网络替代凸优化算法中的正则化程式近算，使得数据项和正则化项可以独立选择并结合。在图像去模糊和去马赛克问题中，通过使用固定的去噪神经网络，本文取得了最先进的重建效果，并且该方法具有很高的通用性和较少的问题特定训练需求。