Filters








268,743 Hits in 6.8 sec

Single Image Restoration for Participating Media Based on Prior Fusion

Joel Felipe de Oliveira Gaya, Amanda Duarte, Felipe Codevilla Moraes, Paulo Drews-Jr, Silvia Silva da Costa Botelho
2019 IEEE Computer Graphics and Applications  
Differently from the related work that only deal with a medium, we obtain generality by using an image formation model and a fusion of new image priors.  ...  The proposed restoration method is based on the fusion of these priors and supported by statistics collected on images acquired in both non-participating and participating media.  ...  In this context, we propose an automatic single image restoration method designed to work in general participative media.  ... 
doi:10.1109/mcg.2018.2881388 fatcat:v2oouryzffca3fwrkdtlrbjy6y

Multi-Prior Learning via Neural Architecture Search for Blind Face Restoration [article]

Yanjiang Yu, Puyang Zhang, Kaihao Zhang, Wenhan Luo, Changsheng Li, Ye Yuan, Guoren Wang
2022 arXiv   pre-print
generates faithful and realistic images.  ...  Blind Face Restoration (BFR) aims to recover high-quality face images from low-quality ones and usually resorts to facial priors for improving restoration performance.  ...  As aforementioned, one single facial prior may be insufficient to generate HQ images.  ... 
arXiv:2206.13962v1 fatcat:mkwpisnkwfahtcfhvq4jciypsq

Image Dehazing in Disproportionate Haze Distributions

Shih-Chia Huang, Da-Wei Jaw, Wen Li Li, Zhihui Lu, Sy-Yen Kuo, Benjamin C. M. Fung, Bo-Hao Chen, Thanisa Numnonda
2021 IEEE Access  
He is a Full Professor with the Department of Electronic Engineering, National Taipei University of Technology, Taipei, and an International Adjunct Professor with the Faculty of Business and Information  ...  maps generated by these two single patch-sizes to produce the hybrid transmission map and restore hazy images with disproportionate haze formation.  ...  s method [14] employs the dark channel prior to produce a transmission map via a single patch-size (e.g., 3 × 3 or 15 × 15) in order to restore a hazy image with disproportionate haze formation, as shown  ... 
doi:10.1109/access.2021.3065968 fatcat:zmxofmuk65bdjftauluaboupdm

Non-blind Deblurring: Handling Kernel Uncertainty with CNNs

Subeesh Vasu, Venkatesh Reddy Maligireddy, A. N. Rajagopalan
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
To generalize the performance to tackle arbitrary kernel noise, we train our network with a large number of real and synthetic noisy blur kernels.  ...  We provide multiple latent image estimates corresponding to different prior strengths obtained from a given blurry observation in order to exploit the complementarity of these inputs for improved learning  ...  The complementarity lies in the fact that restored images with low prior weight preserve details but suffer from artifacts.  ... 
doi:10.1109/cvpr.2018.00345 dblp:conf/cvpr/VasuMR18 fatcat:owmbchycwzaq5hqjfhmqcqjssy

Learning to restore images degraded by atmospheric turbulence using uncertainty [article]

Rajeev Yasarla, Vishal M. Patel
2022 arXiv   pre-print
In this paper, we propose a deep learning-based approach for restring a single image degraded by atmospheric turbulence.  ...  The estimated uncertainty maps are then used to guide the network to obtain the restored image.  ...  In many applications of long-range imaging, such as surveillance, we are faced with a scenario where we have to restore a single image degraded by atmospheric turbulence.  ... 
arXiv:2207.03447v1 fatcat:cp3ssfetc5cihmipd3l53q6x4y

PetsGAN: Rethinking Priors for Single Image Generation [article]

Zicheng Zhang, Yinglu Liu, Congying Han, Hailin Shi, Tiande Guo, Bowen Zhou
2022 arXiv   pre-print
Single image generation (SIG), described as generating diverse samples that have similar visual content with the given single image, is first introduced by SinGAN which builds a pyramid of GANs to progressively  ...  learn the internal patch distribution of the single image.  ...  Related Work Single image generation model. Begin with modeling the internal patch distribution of a single image, SIG models aim to generate new plausible images with high texture quality.  ... 
arXiv:2203.01488v1 fatcat:hviuzkz4ovcmrmnxfv2dtfp4su

Learning to Restore a Single Face Image Degraded by Atmospheric Turbulence using CNNs [article]

Rajeev Yasarla, Vishal M Patel
2020 arXiv   pre-print
We present a deep learning-based solution to the problem of restoring a turbulence-degraded face image where prior information regarding the amount of geometric distortion and blur at each location of  ...  Furthermore, a novel loss is proposed to train TDRN where first and second order image gradients are computed along with their confidence maps to mitigate the effect of turbulence degradation.  ...  The proposed TDRN method is compared with the following recent state-of-the-art face image restoration methods [14] - [16] and generic single image deblurring methods [18] , [19] .  ... 
arXiv:2007.08404v1 fatcat:wky5v3lnzjegpm37poiuax55ya

Deep Image Prior [article]

Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky
2018 arXiv   pre-print
Deep convolutional networks have become a popular tool for image generation and restoration.  ...  Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images.  ...  While in this work we focus on single image restoration, the proposed approach can be extended to the tasks of the restoration of multiple images, e.g. for the task of video restoration.  ... 
arXiv:1711.10925v3 fatcat:i4yhsnlsmbbbdccraq3afjuvli

DeepCFL: Deep Contextual Features Learning from a Single Image [article]

Indra Deep Mastan, Shanmuganathan Raman
2020 arXiv   pre-print
DeepCFL is a single image GAN framework that learns the distribution of the context vectors from the input image.  ...  DeepCFL is applicable when the input source image and the generated target image are not aligned. We illustrate image synthesis using DeepCFL for the task of image resizing.  ...  The generator G learns the context vectors through its interaction with D.  ... 
arXiv:2011.03712v1 fatcat:umqkoosobjcrfltqcj24atrs7i

Deep Image Prior

Victor Lempitsky, Andrea Vedaldi, Dmitry Ulyanov
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Deep convolutional networks have become a popular tool for image generation and restoration.  ...  Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images.  ...  While in this work we focus on single image restoration, the proposed approach can be extended to the tasks of the restoration of multiple images, e.g. for the task of video restoration.  ... 
doi:10.1109/cvpr.2018.00984 dblp:conf/cvpr/UlyanovVL18 fatcat:fmezbrbpvbduxjhifictah5szm

Deep Blind Video Super-resolution [article]

Jinshan Pan, Songsheng Cheng, Jiawei Zhang, Jinhui Tang
2020 arXiv   pre-print
With the estimated blur kernel, we develop an image deconvolution method based on the image formation model of video SR to generate intermediate latent images so that some sharp image contents can be restored  ...  We show that the proposed algorithm is able to generate clearer images with finer structural details.  ...  Intermediate latent image restoration With the blur kernel K, we can estimate HR image from input LR image L i according to (3) . However, solving (3) needs the optical flow and image prior.  ... 
arXiv:2003.04716v1 fatcat:dj5mpfitgjhxpgj6n6vevyvspa

GCFSR: a Generative and Controllable Face Super Resolution Method Without Facial and GAN Priors [article]

Jingwen He, Wu Shi, Kai Chen, Lean Fu, Chao Dong
2022 arXiv   pre-print
Face image super resolution (face hallucination) usually relies on facial priors to restore realistic details and preserve identity information.  ...  In this work, we propose a generative and controllable face SR framework, called GCFSR, which can reconstruct images with faithful identity information without any additional priors.  ...  Existing methods can only output a single restoration result with a fixed style. However, in real scenarios, users might want to adjust the generative strength to meet personalized requirements.  ... 
arXiv:2203.07319v1 fatcat:bwcn7q5flzb2pp4urzoskb36ca

Structural Prior Guided Generative Adversarial Transformers for Low-Light Image Enhancement [article]

Cong Wang and Jinshan Pan and Xiao-Ming Wu
2022 arXiv   pre-print
Our SPGAT mainly contains a generator with two discriminators and a structural prior estimator (SPE).  ...  The generator is based on a U-shaped Transformer which is used to explore non-local information for better clear image restoration.  ...  The generator is used to explore non-local information with the guidance of a structural prior estimator (SPE) for better clear image restoration.  ... 
arXiv:2207.07828v2 fatcat:o66wyz4c5rgcjks532irnjrcjq

Deep Mean-Shift Priors for Image Restoration [article]

Siavash Arjomand Bigdeli, Meiguang Jin, Paolo Favaro, Matthias Zwicker
2017 arXiv   pre-print
We include our prior in a formulation of image restoration as a Bayes estimator that also allows us to solve noise-blind image restoration problems.  ...  In this paper we introduce a natural image prior that directly represents a Gaussian-smoothed version of the natural image distribution.  ...  Super-resolution To demonstrate the generality of our prior, we perform an additional test with single image superresolution.  ... 
arXiv:1709.03749v2 fatcat:7ddmi5sugjbxzmundeitqgg6dq

Good Image Priors for Non-blind Deconvolution [chapter]

Libin Sun, Sunghyun Cho, Jue Wang, James Hays
2014 Lecture Notes in Computer Science  
Most image restoration techniques build "universal" image priors, trained on a variety of scenes, which can guide the restoration of any image.  ...  Re-training generic image priors using ideal sharp example images provides minimal improvement in non-blind deconvolution.  ...  , a known PSF, and one or more sharp images with shared content, how can we reliably remove blur and restore coherent image details?  ... 
doi:10.1007/978-3-319-10593-2_16 fatcat:kqen4kaiyzhorhhcunx6pusxbe
« Previous Showing results 1 — 15 out of 268,743 results