Deep Convolutional Neural Network for Image Deconvolution

By Li Xu, Jimmy S. J. Ren, Ce Liu, Jiaya Jia. 2014

994 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit (2017, methodology): “Quantitatively, we measure the peak signal-to-noise ratio (PSNR; Xu et al. 2014) of the blurred image and the recovered image.”
Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis (2016, background): “It is also possible to re-train networks for specific generative tasks, such as image deblur [28], super-resolution [4], and class visualization [19].”
Generative Image Inpainting with Contextual Attention (2018, background): “Initial efforts [21, 37] train convolutional neural networks for denoising and inpainting of small regions.”
Examining the Impact of Blur on Recognition by Convolutional Networks (2016, background): “Graphics and vision researchers have made significant progress on this problem across the last decade [7, 13, 15, 28]—including with recent neural network-based methods [1, 21, 22, 26].”
Video Frame Interpolation via Adaptive Separable Convolution (2017, methodology): “Our research is inspired by the success of applying deep learning to optical ﬂow estimation [2, 12, 16, 19, 50, 51, 52], artistic style transfer [17, 22, 28], and image enhancement [6, 9, 10, 46, 47, 55, 57, 62, 65].”
Deep joint demosaicking and denoising (2016, background): “…depth and normals estimation from a single image [Eigen et al. 2014; Wang et al. 2015], view interpolation [Flynn et al. 2016], deconvolution [Xu et al. 2014], filter approximation [Xu et al. 2015], image colorization [Cheng et al. 2015; Zhang et al. 2016; Larsson et al. 2016; Iizuka et al.…”
A Sparse-View CT Reconstruction Method Based on Combination of DenseNet and Deconvolution (2018, methodology): “Next, the FBP images were enhanced by an end-to-end neural network that takes full advantage of DenseNet [39] and deconvolution [40] (hereafter the network is referred as DD-Net).”
Using Deep Neural Networks for Inverse Problems in Imaging: Beyond Analytical Methods (2018, background): “[45] (2016) Image SR Multilayer neural network; each layer mimics an operation of the unfolded learned iterative shrinkage and thresholding algorithm (LISTA) for SR”
Grid Search, Random Search, Genetic Algorithm: A Big Comparison for NAS (2019, background): “Over the last few years, convolutional neural networks (CNNs) and their varieties have seen great results on a variety of machine learning problems and applications (Shin et al. (2016); Krizhevsky et al. (2012); Wu (2019); Xu et al. (2014); Goodfellow et al. (2014)).”
Deep Convolutional Neural Network for Inverse Problems in Imaging (2016, background): “Moreover, regression-type neural networks have demonstrated impressive results on inverse problems with exact models such as signal denoising [21], [22], deconvolution [23], artifact reduction [24], [25], signal recovery (modelbased restoration) [26], [27], and interpolation [28]–[30].”
Bidirectional Recurrent Convolutional Networks for Multi-Frame Super-Resolution (2015, background): “Note that for 3D feedforward convolution, varying the number and spatial size of its filters does not have a significant impact on the performance, because some filters with certain spatial sizes are already in a regime where they can already well model the frame content [9], [49].”
Loss Functions for Neural Networks for Image Processing (2015, methodology): “Neural networks have been used for denoising [8,3], deblurring [4], demosaicking [10], and super-resolution [9] among others.”
Fast End-to-End Trainable Guided Filter (2018, background): “Such tasks include image denoising [40], image demosaicking [41], image deblurring [42], image matting [43], rain drop removal [44], image dehazing [45], and image colorization [46].”
From Motion Blur to Motion Flow: A Deep Learning Solution for Removing Heterogeneous Motion Blur (2016, methodology): “Some discriminative methods are proposed for non-blind deconvolution based on Gaussian CRF [30], multi-layer perceptron (MLP) [31], and deep convolution neural network (CNN) [43], etc, which all require the known blur kernels.”
Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections (2016, methodology): “We mainly follow the experimental protocols as in [56] for evaluation of non-blind deblurring.”