A Neural Approach to Blind Motion Deblurring

By Ayan Chakrabarti. European Conference on Computer Vision, 2016

472 citations recorded by the source corpus.

We present a new method for blind motion deblurring that uses a neural network trained to compute estimates of sharp image patches from observations that are blurred by an unknown motion kernel. Instead of regressing directly to patch intensities, this network learns to predict the complex Fourier coefficients of a deconvolution filter to be applied to the input patch for restoration. For inference, we apply the network independently to all overlapping patches in the observed image, and average its outputs to form an initial estimate of the sharp image. We then explicitly estimate a single global blur kernel by relating this estimate to the observed image, and finally perform non-blind deconvolution with this kernel. Our method exhibits accuracy and robustness close to state-of-the-art iterative methods, while being much faster when parallelized on GPU hardware.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/5f2213f1a5caf360d9c7cb29163f7a127135066c

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

In Media Technology

Citations in the corpus

Scale-Recurrent Network for Deep Image Deblurring (2018, methodology): “Data Preparation To create a large training dataset, early learning-based methods [2, 28, 32] synthesize blurred images by convolving sharp images with real or generated uniform/non-uniform blur kernels.”
Rethinking Coarse-to-Fine Approach in Single Image Deblurring (2021, background): “ods [30, 7, 1, 27] commonly exploit CNN as a blur kernel estimator and construct two-stage image deblurring framework, i.”
Classification-Driven Dynamic Image Enhancement (2017, methodology): “Considerable progress has been seen in the development for removing the effects of blur [2], noise [27], and compression artifacts [36] using CNN architectures.”
Deep Image Deblurring: A Survey (2022, background): “ConvolutionalLayer Numerousmethods(Chakrabarti2016; Kaufman and Fattal 2020; Schuler et al. 2015; Sun et al. 2015) train 2D CNNs to directly recover sharp images without kernel estimation steps (Aljadaany et al. 2019; Gong et al. 2017; Lu et al. 2019; Madam Nimisha et al. 2018; Mus-taniemi et al.…”
Finding Tiny Faces in the Wild with Generative Adversarial Network (2018, methodology): “Recently, conventional neural networks have also been used to deblur the blind image [36, 23, 3].”
Neural Blind Deconvolution Using Deep Priors (2019, methodology): “Besides the two regularization terms, we further introduce the non-negative and equality constraints for blur kernel k [29, 32, 41, 56], and the pixels in x are also constrained to the range [0, 1].”
Detecting small faces in the wild based on generative adversarial network and contextual information (2019, methodology): “Recently, CNNs have also been used to deblur the low-quality images [31–34] .”
Deblurring via Stochastic Refinement (2021, methodology): “Alternatively, many recent methods adopt an end-to-end approach where a deep neural network is trained to directly produce a point estimate [8, 11, 14, 21, 35, 36, 50, 54, 55, 64, 66, 67, 73].”
Unsupervised Domain-Specific Deblurring via Disentangled Representations (2019, methodology): “Although numerous blur generation methods have been developed [15, 38, 4], they are not capable of learning all types of blur variants in the wild.”
Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring (2016, methodology): “In [29, 25, 1], synthesized blur images with uniform blur kernel are used for training.”
Learning to Super-Resolve Blurry Face and Text Images (2017, methodology): “Recently, neural networks have also been used for blind image deblurring [39, 3, 37].”
DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks (2017, methodology): “Chakrabarti [6] creates blur kernel by sampling 6 random points and fitting a spline to them.”
Examining the Impact of Blur on Recognition by Convolutional Networks (2016, methodology): “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].”
Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks (2018, background): “Recently, deep learning has been widely used in many low-level vision problems, such as denoising [1, 20, 45], super-resolution[4, 35, 13, 14, 15, 43, 31], dehazing [27], derain/dedirt [6, 5], edge-preserving filtering [39, 18], and image deblurring (non-blind [28, 38, 44] and blind [29, 2, 42]).”
Deblurring by Realistic Blurring (2020, methodology): “Chakrabarti [2] uses a network to compute estimations of sharp images that are blurred by an unknown motion kernel.”
From Motion Blur to Motion Flow: A Deep Learning Solution for Removing Heterogeneous Motion Blur (2016, background): “Chakrabarti [3] trains a patch-based neural network to estimate the frequency information for uniform motion blur removal.”
Deep Video Deblurring for Hand-Held Cameras (2017, methodology): “[4], Kim and Lee [17], Chakrabarti [2], Xu et al.”