Deep joint demosaicking and denoising

Gharbi, Michaël; Chaurasia, Gaurav; Paris, Sylvain; Durand, Frédo

doi:10.1145/2980179.2982399

Cited by 437 publications

(495 citation statements)

References 57 publications

Supporting

Mentioning

468

Contrasting

Unclassified

Order By: Relevance

“…Plug-and-play priors ( [31], [32], [33], [43]): Like the plug-and-play work, our method uses formal optimization with a proximal operator framework. However, while plug-and-play methods adopt an existing generic Gaussian denoiser for the prior proximal operator, our method trains the prior proximal operator with discriminative learning technique.…”

Section: E Connection and Difference With Related Methodsmentioning

confidence: 99%

“…The state-ofthe-art CSF and TRD methods can be derived from the FoE model [30] by unrolling corresponding optimization iterations to be feed-forward networks, where the parameters of each network are trained by minimizing the error between its output images and ground truth for each specific task. Another line of research apply neural networks for image restoration, such as multi-layer perceptrons [40], deep convolutional networks [41], [42], [43] and deep recurrent neural networks [44]. Discriminative approaches owe their computational efficiency at run-time to a particular feed-forward structure whose trainable parameters are optimized for a particular task during training.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discriminative Transfer Learning for General Image Restoration

Xiao

Heide

Heidrich

et al. 2018

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Abstract-Recently, several discriminative learning approaches have been proposed for effective image restoration, achieving convincing trade-off between image quality and computational efficiency. However, these methods require separate training for each restoration task (e.g., denoising, deblurring, demosaicing) and problem condition (e.g., noise level of input images). This makes it time-consuming and difficult to encompass all tasks and conditions during training. In this paper, we propose a discriminative transfer learning method that incorporates formal proximal optimization and discriminative learning for general image restoration. The method requires a single-pass discriminative training and allows for reuse across various problems and conditions while achieving an efficiency comparable to previous discriminative approaches. Furthermore, after being trained, our model can be easily transferred to new likelihood terms to solve untrained tasks, or be combined with existing priors to further improve image restoration quality.

show abstract

Section: E Connection and Difference With Related Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Discriminative Transfer Learning for General Image Restoration

Xiao

Heide

Heidrich

et al. 2018

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…Wang [Wan14] used 4 × 4 patches to train a multilayer neural network while minimizing a suitable objective function. Gharbi et al constructed a dataset with hard cases, which were used to train a CNN for joint demosaicing and denoising [GCPD16]. All these methods were specifically designed to reconstruct Bayer filtered images.…”

Section: Related Workmentioning

confidence: 99%

Deep Joint Design of Color Filter Arrays and Demosaicing

Henz

Gastal

Oliveira

2018

Computer Graphics Forum

View full text Add to dashboard Cite

We present a convolutional neural network architecture for performing joint design of color filter array (CFA) patterns and demosaicing. Our generic model allows the training of CFAs of arbitrary sizes, optimizing each color filter over the entire RGB color space. The patterns and algorithms produced by our method provide high‐quality color reconstructions. We demonstrate the effectiveness of our approach by showing that its results achieve higher PSNR than the ones obtained with state‐of‐the‐art techniques on all standard demosaicing datasets, both for noise‐free and noisy scenarios. Our method can also be used to obtain demosaicing strategies for pre‐defined CFAs, such as the Bayer pattern, for which our results also surpass even the demosaicing algorithms specifically designed for such a pattern.

show abstract

“…Improving the performance of classifiers and CNNs by augmenting training data sets is widely known and well established [34][35][36][37]. Common data augmentation methods are to shift, rotate, scale, flip, crop, transform, compress or blur the training images to extend the training database.…”

Section: Related Workmentioning

confidence: 99%

UAS Navigation with SqueezePoseNet—Accuracy Boosting for Pose Regression by Data Augmentation

Mueller

Jutzi

2018

Drones

View full text Add to dashboard Cite

Abstract:The navigation of Unmanned Aerial Vehicles (UAVs) nowadays is mostly based on Global Navigation Satellite Systems (GNSSs). Drawbacks of satellite-based navigation are failures caused by occlusions or multi-path interferences. Therefore, alternative methods have been developed in recent years. Visual navigation methods such as Visual Odometry (VO) or visual Simultaneous Localization and Mapping (SLAM) aid global navigation solutions by closing trajectory gaps or performing loop closures. However, if the trajectory estimation is interrupted or not available, a re-localization is mandatory. Furthermore, the latest research has shown promising results on pose regression in 6 Degrees of Freedom (DoF) based on Convolutional Neural Networks (CNNs). Additionally, existing navigation methods can benefit from these networks. In this article, a method for GNSS-free and fast image-based pose regression by utilizing a small Convolutional Neural Network is presented. Therefore, a small CNN (SqueezePoseNet) is utilized, transfer learning is applied and the network is tuned for pose regression. Furthermore, recent drawbacks are overcome by applying data augmentation on a training dataset utilizing simulated images. Experiments with small CNNs show promising results for GNSS-free and fast localization compared to larger networks. By training a CNN with an extended data set including simulated images, the accuracy on pose regression is improved up to 61.7% for position and up to 76.0% for rotation compared to training on a standard not-augmented data set.

show abstract

Deep joint demosaicking and denoising

Cited by 437 publications

References 57 publications

Discriminative Transfer Learning for General Image Restoration

Discriminative Transfer Learning for General Image Restoration

Deep Joint Design of Color Filter Arrays and Demosaicing

UAS Navigation with SqueezePoseNet—Accuracy Boosting for Pose Regression by Data Augmentation

Contact Info

Product

Resources

About