Filters








1,015 Hits in 2.4 sec

Discriminative Non-blind Deblurring

Uwe Schmidt, Carsten Rother, Sebastian Nowozin, Jeremy Jancsary, Stefan Roth
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
Motivation • Non-blind deblurring is an important component for removing image blur (e.g. due to camera shake) after blur estimation. • High-quality learning-based methods have been limited to the generative  ...  case and are often computationally expensive. • Hand-defined models with inferior quality are most widely used. • How to devise a flexible discriminative approach with high restoration quality and efficiency  ...  Our Approach • First discriminative approach for non-blind deblurring for arbitrary images and blurs. • Efficient with state-of-the-art results on three benchmarks. • Generalizes commonly-used half-quadratic  ... 
doi:10.1109/cvpr.2013.84 dblp:conf/cvpr/SchmidtRNJR13 fatcat:twjhqp2cfnalvpw7wle3xduxme

SL-CycleGAN: Blind Motion Deblurring in Cycles using Sparse Learning [article]

Ali Syed Saqlain, Li-Yun Wang, Fang Fang
2021 arXiv   pre-print
Memory) to replace non-linearity such as ReLU in the ResNet-block of SL-CycleGAN generators.  ...  In this paper, we introduce an end-to-end generative adversarial network (GAN) based on sparse learning for single image blind motion deblurring, which we called SL-CycleGAN.  ...  To address the problem of non-uniform blind image deblurring, and to propose such a GAN-based blind image motion deblurring network, that, unlike other GAN-based models does not treat the restoration of  ... 
arXiv:2111.04026v1 fatcat:phvtm3b53rhdppsnewmntf6vme

Learning a Discriminative Prior for Blind Image Deblurring [article]

Lerenhan Li, Jinshan Pan, Wei-Sheng Lai, Changxin Gao, Nong Sang, Ming-Hsuan Yang
2018 arXiv   pre-print
We present an effective blind image deblurring method based on a data-driven discriminative prior.Our work is motivated by the fact that a good image prior should favor clear images over blurred images.In  ...  deblurring in various scenarios, including natural, face, text, and low-illumination images.However, it is difficult to optimize the deblurring method with the learned image prior as it involves a non-linear  ...  Extension to Non-Uniform Deblurring The proposed discriminative image prior can be easily extended for non-uniform motion deblurring.  ... 
arXiv:1803.03363v2 fatcat:kqt3gsiygbemtjznljru373o6q

Interleaved Regression Tree Field Cascades for Blind Image Deconvolution

Kevin Schelten, Sebastian Nowozin, Jeremy Jancsary, Carsten Rother, Stefan Roth
2015 2015 IEEE Winter Conference on Applications of Computer Vision  
using a non-blind deblurring algorithm.  ...  Recent work in non-blind deblurring has shown that discriminative approaches can have clear image quality and runtime benefits over typical generative formulations.  ...  This generalizes previous work on RTF cascades for non-blind deblurring [21] to the blind deblurring task.  ... 
doi:10.1109/wacv.2015.72 dblp:conf/wacv/ScheltenNJRR15 fatcat:ggwlxxqdenbvtoju2busw4ilmy

Learning a Discriminative Prior for Blind Image Deblurring

Lerenhan Li, Jinshan Pan, Wei-Sheng Lai, Changxin Gao, Nong Sang, Ming-Hsuan Yang
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We present an effective blind image deblurring method based on a data-driven discriminative prior.  ...  Furthermore, the proposed model can be easily extended to non-uniform deblurring.  ...  Extension to Non-Uniform Deblurring The proposed discriminative image prior can be easily extended for non-uniform motion deblurring.  ... 
doi:10.1109/cvpr.2018.00692 dblp:conf/cvpr/LiPLGS018 fatcat:lrwf32pqqvfihihxukc6hix6em

FMD-cGAN: Fast Motion Deblurring using Conditional Generative Adversarial Networks [article]

Jatin Kumar and Indra Deep Mastan and Shanmuganathan Raman
2021 arXiv   pre-print
In this paper, we present a Fast Motion Deblurring-Conditional Generative Adversarial Network (FMD-cGAN) that helps in blind motion deblurring of a single image.  ...  The resulting compressed Deblurring cGAN faster than its closest competitors and even qualitative and quantitative results outperform various recently proposed state-of-the-art blind motion deblurring  ...  We have described that image deblurring is classified into two types: Non-blind image deblurring and Blind image deblurring (Sec. 1).  ... 
arXiv:2111.15438v2 fatcat:ouphc37dp5avnhxrrvgx47dbge

Cascades of Regression Tree Fields for Image Restoration

Uwe Schmidt, Jeremy Jancsary, Sebastian Nowozin, Stefan Roth, Carsten Rother
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Our experiments show that when applied to non-blind image deblurring, the proposed approach is efficient and yields state-of-the-art restoration quality on images corrupted with synthetic and real blur  ...  For image deblurring, however, discriminative approaches have been mostly lacking.  ...  ACKNOWLEDGMENTS We thank Pushmeet Kohli for suggesting the topic of discriminative deblurring using a non-parametric model like the RTF; we also thank Stephan Richter for help in preparation of  ... 
doi:10.1109/tpami.2015.2441053 pmid:26959673 fatcat:3qfldh5asne73krix3mfx225va

Learning Discriminative Data Fitting Functions for Blind Image Deblurring

Jinshan Pan, Jiangxin Dong, Yu-Wing Tai, Zhixun Su, Ming-Hsuan Yang
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
Solving blind image deblurring usually requires defining a data fitting function and image priors.  ...  While existing algorithms mainly focus on developing image priors for blur kernel estimation and non-blind deconvolution, only a few methods consider the effect of data fitting functions.  ...  Discriminative Non-Blind Deconvolution Once blur kernels are obtained, we can use a variety of non-blind deconvolution methods to recover latent images.  ... 
doi:10.1109/iccv.2017.122 dblp:conf/iccv/PanDTS017 fatcat:hkksxv6ifnb6fkgi46icomggim

Blur Invariant Kernel-Adaptive Network for Single Image Blind deblurring [article]

Sungkwon An, Hyungmin Roh, Myungjoo Kang
2020 arXiv   pre-print
We present a novel, blind, single image deblurring method that utilizes information regarding blur kernels.  ...  Subsequently, we propose a deblurring network that restores sharp images using the estimated blur kernel.  ...  RELATED STUDIES In this section, we provide a brief review of some non-blind and blind approaches for solving the image deblurring problem.  ... 
arXiv:2007.04543v3 fatcat:mhyenqhhjnd4roap6tzqkmo5za

GAN-based image deblurring using DCT loss with customized datasets

Hiroki Tomosada, Takahiro Kudo, Takanori Fujisawa, Masaaki Ikehara
2021 IEEE Access  
Experimented code with pre-trained weights, datasets and results are available at https://github.com/Hiroki-Tomosada/DCTGAN-master INDEX TERMS Image deblurring, Blind deconvolution, Non-Uniform, GAN, DCT  ...  With this loss, DeblurDCTGAN can reduce block noise and ringing artifacts while maintaining deblurring performance.  ...  Image deconvolution is classified into two types of approaches: non-blind and blind.  ... 
doi:10.1109/access.2021.3116194 fatcat:ck2y2svfqzax7newy3nmg7zvtu

Motion Blur Image Restoration by Multi-Scale Residual Neural Network

Xu Hexin, Zhao Li, Jiao Yan
2021 International Journal of Advanced Network, Monitoring, and Controls  
Blind deblurring is a basic subject of computer vision and image processing. Motion image deblurring is divided into non blind deblurring and blind deblurring by whether to estimate the blur kernel.  ...  Blind deblurring is easy to produce motion artifacts because of the inaccurate estimation of the blur kernel. Non blind deblurring is the best choice for the current blurred image processing.  ...  According to the nature of the blur kernel, it is divided into blind deblurring and non-blind deblurring.  ... 
doi:10.21307/ijanmc-2021-009 fatcat:66unjkckqfdphjzra4qbw6wbsq

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks [article]

Orest Kupyn, Volodymyr Budzan, Mykola Mykhailych, Dmytro Mishkin, Jiri Matas
2018 arXiv   pre-print
We present DeblurGAN, an end-to-end learned method for motion deblurring. The learning is based on a conditional GAN and the content loss .  ...  The quality of the deblurring model is also evaluated in a novel way on a real-world problem -- object detection on (de-)blurred images.  ...  The family of deblurring problems is divided into two types: blind and non-blind deblurring.  ... 
arXiv:1711.07064v4 fatcat:nzcsivvhyjcn5jhsqswablshm4

Blind Motion Deblurring through SinGAN Architecture [article]

Harshil Jain, Rohit Patil, Indra Deep Mastan, Shanmuganathan Raman
2020 arXiv   pre-print
In this paper, we focus on blind motion deblurring through SinGAN architecture.  ...  Blind motion deblurring involves reconstructing a sharp image from an observation that is blurry. It is a problem that is ill-posed and lies in the categories of image restoration problems.  ...  Non-blind deblurring is based on the assumption that the blur kernels kernel(M ) are known.  ... 
arXiv:2011.03705v1 fatcat:tbeaiakd5vd3bar3556qnrba5a

Generative Adversarial Network Based on Multi-feature Fusion Strategy for Motion Image Deblurring

Zhou-xiang Jin Zhou-xiang Jin, Hao Qin Zhou-xiang Jin
2022 Diànnǎo xuékān  
Image deblurring can be divided into two categories: one is the non-blind image deblurring with known fuzzy kernel, and the other is the blind image deblurring with unknown fuzzy kernel.  ...  The traditional motion image deblurring networks ignore the non-uniformity of motion blurred images and cannot effectively recover the high frequency details and remove artifacts.  ...  References [5, 6] used convolutional neural networks to estimate the blur kernel, and then used traditional non-blind deblurring algorithms to obtain clear images in motion blur.  ... 
doi:10.53106/199115992022023301004 fatcat:qprx2szja5dtfjlt67mljqxuc4

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

Orest Kupyn, Volodymyr Budzan, Mykola Mykhailych, Dmytro Mishkin, Jiri Matas
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
The method is 5 times faster than the closest competitor -Deep-Deblur [25] .  ...  We present DeblurGAN, an end-to-end learned method for motion deblurring. The learning is based on a conditional GAN and the content loss .  ...  The family of deblurring problems is divided into two types: blind and non-blind deblurring.  ... 
doi:10.1109/cvpr.2018.00854 dblp:conf/cvpr/KupynBMMM18 fatcat:ewapgi3ti5cm7nhszp2zkemoqu
« Previous Showing results 1 — 15 out of 1,015 results