Learning High-Order Filters for Efficient Blind Deconvolution of Document Photographs

Xiao, Lei; Wang, Jue; Heidrich, Wolfgang; Hirsch, Michael

doi:10.1007/978-3-319-46487-9_45

Cited by 41 publications

(21 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3) Iterative shrinkage: This is a variant of sparse reconstruction which recasts the regularization problem as an iterative procedure where dominant feature coefficients are preserved during each iteration. Different regularizers can be found for image deconvolution in [57]- [61]. 4) Variational regularization: Known as the total variation (TV) method, in which the priors for either the blur kernel or latent image are regulated by the TV-norm [48]- [52], this norm preserves sharp edges while preventing Gibbs oscillations for recovery.…”

Section: Arxiv:181010725v2 [Eessiv] 19 Jul 2019mentioning

confidence: 99%

“…Trained CNN model to classify blur vs clean as image prior for regularized minimization formulation Simoes [35] 2016 NB • Diagonalizing unknown convolution operator using FFT and solving via ADMM Kim [36] 2015 B • • Encode temporal/spatial coherency of dynamic scene using optical-flow/TV regularized minimization Liu [37] 2014 B • • • Estimate blur from image spectral property and feed into a regularized TV/eigenvalue minimization Mosleh [38] 2014 N • • • Encode ringing artifacts using Gabor wavelets and fit into a regularized minimization for cancelation Pan [39] 2014 N • • • Text image deblurring regularized by sparse encoding of spatial/gradient domains Pan [40] 2013 B • • Estimates the kernel and the deblurred image from a combined sparse regularization framework Kim [41] 2013 B • • • Dynamic image deconvolution using TV/Tikhonov/temporal-sparsity regularized minimization Shen [42] 2012 B • • • TV/Tikhonov regularized minimization for image deconvolution Sroubek [43] 2012 B • • 1-regularized minimization for image deconvolution Dong [44] 2011 NB • • Learn adaptive bases and use in adaptive regularized minimization for sparse reconstruction Zhang [45], [46] 2011 B • • • Sparse regulation of images via KSVD library for deconvolution and apply to facial recognition Bai [47] 2018 B • • Both kernel/image recovered via combined regularization using reweighted graph TV priors Lou [48] 2015 N • • Weighted differences of TV regularizers in 1/ 2 norms and solved by split variable technique Zhang [49] 2014 N • • • Local/non-local similarities defined by TV 1 /TV 2 and regulated by combined minimization Xu [50] 2012 B • Regulate motion by difference of depth map and deconvolve via non-convex TV minimization Chan [51] 2011 N • Deconvolve image/videos using spatial/temporal TV regularization solved by split varying technique Afonso [52] 2010 N • Deconvolve image using TV regularization solved by split varying technique Li [53] 2018 B • • Non-iterative deconvolution via combination of Wiener filters, solution by a system linear equations Bertero [54] 2010 N • Generalized Kullback-Leiblar divergence function to regularize Poisson images Cho [16] 2009 B • • Separate recovery of motion kernel and image from residual image using Tikhonov regularization Wiener [55], [56] 1949 N • Regulate image spectrum in Fourier domain with inverse kernel response Xiao …”

Section: Authormentioning

confidence: 99%

See 1 more Smart Citation

Convolutional Deblurring for Natural Imaging

Hosseini

Plataniotis

2020

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

In this paper, we propose a novel design of image deblurring in the form of one-shot convolution filtering that can directly convolve with naturally blurred images for restoration. The problem of optical blurring is a common disadvantage to many imaging applications that suffer from optical imperfections. Despite numerous deconvolution methods that blindly estimate blurring in either inclusive or exclusive forms, they are practically challenging due to high computational cost and low image reconstruction quality. Both conditions of high accuracy and high speed are prerequisites for high-throughput imaging platforms in digital archiving. In such platforms, deblurring is required after image acquisition before being stored, previewed, or processed for high-level interpretation. Therefore, on-the-fly correction of such images is important to avoid possible time delays, mitigate computational expenses, and increase image perception quality. We bridge this gap by synthesizing a deconvolution kernel as a linear combination of Finite Impulse Response (FIR) evenderivative filters that can be directly convolved with blurry input images to boost the frequency fall-off of the Point Spread Function (PSF) associated with the optical blur. We employ a Gaussian low-pass filter to decouple the image denoising problem for image edge deblurring. Furthermore, we propose a blind approach to estimate the PSF statistics for two Gaussian and Laplacian models that are common in many imaging pipelines. Thorough experiments are designed to test and validate the efficiency of the proposed method using 2054 naturally blurred images across six imaging applications and seven state-of-the-art deconvolution methods.

show abstract

Section: Arxiv:181010725v2 [Eessiv] 19 Jul 2019mentioning

confidence: 99%

Section: Authormentioning

confidence: 99%

Convolutional Deblurring for Natural Imaging

Hosseini

Plataniotis

2020

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…This approach has been extended to license plates in [36]. [40] proposes to learn a multi-scale cascade of shrinkage fields model. This model however does not seem to generalize to natural images.…”

Section: Related Workmentioning

confidence: 99%

Motion Deblurring in the Wild

Noroozi

Chandramouli

Favaro

2017

Lecture Notes in Computer Science

118

View full text Add to dashboard Cite

We propose a deep learning approach to remove motion blur from a single image captured in the wild, i.e., in an uncontrolled setting. Thus, we consider motion blur degradations that are due to both camera and object motion, and by occlusion and coming into view of objects. In this scenario, a modelbased approach would require a very large set of parameters, whose fitting is a challenge on its own. Hence, we take a data-driven approach and design both a novel convolutional neural network architecture and a dataset for blurry images with ground truth. The network produces directly the sharp image as output and is built into three pyramid stages, which allow to remove blur gradually from a small amount, at the lowest scale, to the full amount, at the scale of the input image. To obtain corresponding blurry and sharp image pairs, we use videos from a high frame-rate video camera. For each small video clip we select the central frame as the sharp image and use the frame average as the corresponding blurred image. Finally, to ensure that the averaging process is a sufficient approximation to real blurry images we estimate optical flow and select frames with pixel displacements smaller than a pixel. We demonstrate state of the art performance on datasets with both synthetic and real images.

show abstract

“…Some representative examples of such methods include trainable random field models such as separable Markov random field (MRFSepa) [35] regression tree fields (RTF) [36], cascaded shrinkage fields (CSF) [8], trainable nonlinear reaction diffusion (TRD) models [9] and their extensions [37], [38], [39]. 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.…”

Section: Introductionmentioning

confidence: 99%

Discriminative Transfer Learning for General Image Restoration

Xiao

Heide

Heidrich

et al. 2018

IEEE Trans. on Image Process.

Self Cite

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

Learning High-Order Filters for Efficient Blind Deconvolution of Document Photographs

Cited by 41 publications

References 20 publications

Convolutional Deblurring for Natural Imaging

Convolutional Deblurring for Natural Imaging

Motion Deblurring in the Wild

Discriminative Transfer Learning for General Image Restoration

Contact Info

Product

Resources

About