539,703 Hits in 3.1 sec

Deep Image Prior

Victor Lempitsky, Andrea Vedaldi, Dmitry Ulyanov
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Furthermore, the same prior can be used to invert deep neural representations to diagnose them, and to restore images based on flash-no flash input pairs.  ...  It also bridges the gap between two very popular families of image restoration methods: learning-based methods using deep convolutional networks and learning-free methods based on handcrafted image priors  ...  Pre-images are found either Inversion with deep image prior Inversion with TV prior [21] Pre-trained deep inverting network [8] Figure 9 : AlexNet inversion.  ... 
doi:10.1109/cvpr.2018.00984 dblp:conf/cvpr/UlyanovVL18 fatcat:fmezbrbpvbduxjhifictah5szm

Deep Image Prior [article]

Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky
2018 arXiv   pre-print
Furthermore, the same prior can be used to invert deep neural representations to diagnose them, and to restore images based on flash-no flash input pairs.  ...  It also bridges the gap between two very popular families of image restoration methods: learning-based methods using deep convolutional networks and learning-free methods based on handcrafted image priors  ...  Such behaviour highlights the relation between the deep image prior and traditional self-similarity priors. In fig. 8 , we compare deep priors corresponding to several architectures.  ... 
arXiv:1711.10925v3 fatcat:i4yhsnlsmbbbdccraq3afjuvli

Unsupervised Image Fusion Using Deep Image Priors [article]

Xudong Ma, Paul Hill, Nantheera Anantrasirichai, Alin Achim
2022 arXiv   pre-print
Deep Image Prior (DIP) has been introduced to exploit convolutional neural networks' ability to synthesize the 'prior' in the input image.  ...  A significant number of researchers have applied deep learning methods to image fusion.  ...  This paper proposes a new technique, based on Deep Image Prior (DIP) [18] , to fuse images in an unsupervised way.  ... 
arXiv:2110.09490v3 fatcat:utmas3ebrnc37jh5gromfm36d4

Blind Image Deconvolution using Deep Generative Priors [article]

Muhammad Asim, Fahad Shamshad, Ali Ahmed
2019 arXiv   pre-print
This paper proposes a novel approach to regularize the ill-posed and non-linear blind image deconvolution (blind deblurring) using deep generative networks as priors.  ...  To address the shortcomings of generative models such as mode collapse, we augment our generative priors with classical image priors and report improved performance on complex image datasets.  ...  image deconvolution (blind deblurring) using deep generative networks as priors.  ... 
arXiv:1802.04073v4 fatcat:cuj2bsznhvfkzdtf4q6o5dmepq

Dual Image Deblurring Using Deep Image Prior

Chang Jong Shin, Tae Bok Lee, Yong Seok Heo
2021 Electronics  
Recently, deep image prior (DIP) has shown that convolutional neural networks (CNNs) can be a powerful prior for a single natural image.  ...  Blind image deblurring, one of the main problems in image restoration, is a challenging, ill-posed problem. Hence, it is important to design a prior to solve it.  ...  [17] proposed a deep image prior (DIP), which is based on self-supervised learning, and showed that a CNN can capture the low-level statistics of a single natural image.  ... 
doi:10.3390/electronics10172045 fatcat:wv7xu7hbwzhu7h6a75f7owajw4

DeepRED: Deep Image Prior Powered by RED [article]

Gary Mataev, Michael Elad, Peyman Milanfar
2019 arXiv   pre-print
One such contribution, which is the focus of this paper, is the Deep Image Prior (DIP) work by Ulyanov, Vedaldi, and Lempitsky (2018).  ...  In this work, we aim to boost DIP by adding an explicit prior, which enriches the overall regularization effect in order to lead to better-recovered images.  ...  This work presents the Deep Image Prior (DIP) method, a new strategy for handling the regularization task in inverse problems.  ... 
arXiv:1903.10176v3 fatcat:fr7xp66pevgvvclvqunh4zmr7m

Weak deep priors for seismic imaging [article]

Ali Siahkoohi, Gabrio Rizzuti, Felix J. Herrmann
2021 arXiv   pre-print
This is the so-called deep-prior approach. In seismic imaging, however, evaluating the forward operator is computationally expensive, and training a randomly initialized CNN becomes infeasible.  ...  Our synthetic numerical experiments demonstrate that the weak deep prior is more robust with respect to noise than conventional least-squares imaging approaches, with roughly twice the computational cost  ...  We propose the weak deep prior, a computationally convenient formulation that relaxes deep priors.  ... 
arXiv:2004.06835v2 fatcat:qfx35knjtff7fcwzhermfmqqne

Early Stopping for Deep Image Prior [article]

Hengkang Wang, Taihui Li, Zhong Zhuang, Tiancong Chen, Hengyue Liang, Ju Sun
2022 arXiv   pre-print
Deep image prior (DIP) and its variants have showed remarkable potential for solving inverse problems in computer vision, without any extra training data.  ...  Table 1 1 List of Common Acronyms (in alphabetic order) CNN convolutional neural network DD deep decoder DIP deep image prior DIP-TV DIP with total variation regularization DL deep learning DNN deep neural  ...  Deep image prior (DIP) The idea (Ulyanov, Vedaldi, & Lempitsky, 2018) is simple: parameterize x as x = G θ (z), where G θ is a trainable DNN parametrized by θ and z is a trainable or frozen random seed  ... 
arXiv:2112.06074v2 fatcat:sjijoe7rsnchrg7bgbee2vihh4

Rethinking Deep Image Prior for Denoising [article]

Yeonsik Jo, Se Young Chun, Jonghyun Choi
2021 arXiv   pre-print
Deep image prior (DIP) serves as a good inductive bias for diverse inverse problems.  ...  Our empirical validations show that given a single noisy image, our method denoises the image while preserving rich textual details.  ...  Each prior assumes that the prior distribution is smoothness, low rank and self-similarity, respectively. Image prior by deep neural networks. Ulyanov et al.  ... 
arXiv:2108.12841v1 fatcat:bexeypgirbfwxa3oubd2voga4y

Compressive Ptychography using Deep Image and Generative Priors [article]

Semih Barutcu, Doğa Gürsoy, Aggelos K. Katsaggelos
2022 arXiv   pre-print
To address this bottleneck, we propose a generative model combining deep image priors with deep generative priors.  ...  Ptychography is a well-established coherent diffraction imaging technique that enables non-invasive imaging of samples at a nanometer scale.  ...  We pose the phase retrieval as a compressed sensing problem and focus on combining two state-of-the-art methods: deep generative priors and deep image priors. 1) Deep Generative Priors: Generative Adversarial  ... 
arXiv:2205.02397v3 fatcat:g7zhzszu2jbnhgylibcpdn3f7a

Deep Networks for Image Super-Resolution with Sparse Prior [article]

Zhaowen Wang, Ding Liu, Jianchao Yang, Wei Han, Thomas Huang
2015 arXiv   pre-print
Deep learning techniques have been successfully applied in many areas of computer vision, including low-level image restoration problems.  ...  For image super-resolution, several models based on deep neural networks have been recently proposed and attained superior performance that overshadows all previous handcrafted models.  ...  Conclusions We propose a new model for image SR by combining the strengths of sparse coding and deep network, and make considerable improvement over existing deep and shallow SR models both quantitatively  ... 
arXiv:1507.08905v4 fatcat:z3qebqxoorbmngybn2k55ldqwu

Image Denoising Using Nonlocal Regularized Deep Image Prior

Zhonghua Xie, Lingjun Liu, Zhongliang Luo, Jianfeng Huang
2021 Symmetry  
Specifically, we propose a deep-learning-based method based on the deep image prior (DIP) method, which only requires a noisy image as training data, without any clean data.  ...  Deep neural networks have shown great potential in various low-level vision tasks, leading to several state-of-the-art image denoising techniques.  ...  Conclusions We have proposed an effective iterative algorithm equipped with the deep image prior and the plug-and-play nonlocal priors for image denoising.  ... 
doi:10.3390/sym13112114 fatcat:bdvu6fpeszb2vhquc7clpggx3i

Image Deconvolution with Deep Image and Kernel Priors [article]

Zhunxuan Wang, Zipei Wang, Qiqi Li, Hakan Bilen
2019 arXiv   pre-print
On the success of the recently proposed deep image prior (DIP), we build an image deconvolution model with deep image and kernel priors (DIKP).  ...  Instead, our DIKP model uses such priors in image deconvolution to model not only images but also kernels, combining the ideas of traditional learning-free deconvolution methods with neural nets.  ...  In the rest of the report, f (θ) denotes output image by deep image prior f with weight θ.  ... 
arXiv:1910.08386v1 fatcat:ekvjjt4e5bg25dtgrclix3edde

Blind Image Deconvolution Using Variational Deep Image Prior [article]

Dong Huo, Abbas Masoumzadeh, Rafsanjany Kushol, Yee-Hong Yang
2022 arXiv   pre-print
This paper proposes a new variational deep image prior (VDIP) for blind image deconvolution, which exploits additive hand-crafted image priors on latent sharp images and approximates a distribution for  ...  Deep image prior (DIP) provides an approach to optimize the weights of a randomly initialized network with a single degraded image by maximum a posteriori (MAP), which shows that the architecture of a  ...  None of the mentioned deep-learning-based methods consider the standard deviation of the image. Deep Image Prior Ulyanov et al.  ... 
arXiv:2202.00179v1 fatcat:me7dlctpbrfk7djmfuqnggf7xm

Deep Image Super Resolution via Natural Image Priors [article]

Hojjat S. Mousavi, Tiantong Guo, Vishal Monga
2018 arXiv   pre-print
Experimental results show that the proposed deep network with natural image priors is particularly effective in training starved regimes.  ...  We propose to regularize deep structures with prior knowledge about the images so that they can capture more structural information from the same limited data.  ...  To overcome this problem, we develop a novel deep network that is regularized with prior knowledge of images (natural image priors).  ... 
arXiv:1802.02721v1 fatcat:7qcxh6oafvbrdnt5mesqk5igvq
« Previous Showing results 1 — 15 out of 539,703 results