10,412 Hits in 8.0 sec

Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution [article]

Jie Liang and Hui Zeng and Lei Zhang
2022 arXiv   pre-print
Efficient and effective real-world image super-resolution (Real-ISR) is a challenging task due to the unknown complex degradation of real-world images and the limited computation resources in practical  ...  In this paper, we propose an efficient and effective degradation-adaptive super-resolution (DASR) network, whose parameters are adaptively specified by estimating the degradation of each input image.  ...  Conclusion In this paper, we proposed an efficient degradation-adaptive network, namely DASR, for the real-world image super-resolution (Real-ISR) task.  ... 
arXiv:2203.14216v1 fatcat:ljrevrmj75d5ljqexd5hs2svvq

Editorial: Introduction to the Issue on Deep Learning for Image/Video Restoration and Compression

A. Murat Tekalp, Michele Covell, Radu Timofte, Chao Dong
2021 IEEE Journal on Selected Topics in Signal Processing  
Also In "Multi-scale image super-resolution via a single extendable deep network" Zhang et al. propose a solution (MSWSR) addressing efficiency and arbitrary upscaling factors.  ...  data drops sharply on real-world images/video, where the quantity and quality of training data is limited, and v) exploiting temporal correlations for efficient and effective video restoration and compression  ... 
doi:10.1109/jstsp.2021.3053364 fatcat:hjo5pvw6lvgpfga2wfq4vpaq3q

A Review on Self Learning based Methods for Real World Single Image Super Resolution [chapter]

Yogesh J. Gaikwad
2021 New Frontiers in Communication and Intelligent Systems  
One of the active research issues in image processing is super resolution, which is used to boost picture resolution.  ...  The super resolution of a single image is obtained by rebuilding high-resolution (HR) pictures from low-resolution (LR) damaged pho tos (RSISR).  ...  As there are two major limitations of real-world images suffers from degradation problem therefore it is necessary to adapt RSISR models with ever changing real-world images.  ... 
doi:10.52458/978-81-95502-00-4-1 fatcat:wxz4mp33wzfvhafjx6plw5rnky

Toward Real-world Image Super-resolution via Hardware-based Adaptive Degradation Models [article]

Rui Ma, Johnathan Czernik, Xian Du
2021 arXiv   pre-print
Most single image super-resolution (SR) methods are developed on synthetic low-resolution (LR) and high-resolution (HR) image pairs, which are simulated by a predetermined degradation operation, e.g.,  ...  We design an adaptive blurring layer (ABL) in the supervised learning framework to estimate the target LR images. The hyperparameters of the ABL can be adjusted for different imaging hardware.  ...  Experimental Setups Dataset SupER dataset is a newly proposed high-quality paired HR and LR image dataset for real-world image super-resolution tasks.  ... 
arXiv:2110.10755v1 fatcat:m42bndrminhdpcsktrjv7qt3pa

Real-Time Super-Resolution for Real-World Images on Mobile Devices [article]

Jie Cai, Zibo Meng, Jiaming Ding, Chiu Man Ho
2022 arXiv   pre-print
Extensive experiments on traditional super-resolution datasets (Set5, Set14, BSD100, Urban100, Manga109, DIV2K) and real-world images with a variety of degradations demonstrate that our method outperforms  ...  In this work, an approach for real-time ISR on mobile devices is presented, which is able to deal with a wide range of degradations in real-world scenarios.  ...  The AIM 2019 Challenge on Real-World Image Super-Resolution [23] and the NTIRE 2020 Challenge on Real-World Image Super-Resolution [22] aim to stimulate research in the direction of real-world ISR,  ... 
arXiv:2206.01777v1 fatcat:itcigackyrc2tpemwnsadqfimy

Real-World Single Image Super-Resolution: A Brief Review [article]

Honggang Chen, Xiaohai He, Linbo Qing, Yuanyuan Wu, Chao Ren, Ce Zhu
2021 arXiv   pre-print
This article aims to make a comprehensive review on real-world single image super-resolution (RSISR).  ...  Recent studies show that simulation results on synthetic data usually overestimate the capacity to super-resolve real-world images.  ...  ., realistic datasets for model training and testing, specific models for real-world image super-resolution and reconstruction performance evaluation.  ... 
arXiv:2103.02368v1 fatcat:fesf6ercffdxdg7jx3mda6nkri

Learning Multiple Probabilistic Degradation Generators for Unsupervised Real World Image Super Resolution [article]

Sangyun Lee, Sewoong Ahn, Kwangjin Yoon
2022 arXiv   pre-print
Unsupervised real world super resolution (USR) aims at restoring high-resolution (HR) images given low-resolution (LR) inputs when paired data is unavailable.  ...  Instead, we propose the probabilistic degradation generator. Our degradation generator is a deep hierarchical latent variable model and more suitable for modeling the complex distribution.  ...  GAN-based Unsupervised Real-World Super Resolution Although SR models achieved remarkable results on bicubicdownsampled data, they perform poorly in the real world.  ... 
arXiv:2201.10747v1 fatcat:sl4wa5quyjelboxr7nr6y2tfp4

Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution

Yong Guo, Jian Chen, Jingdong Wang, Qi Chen, Jiezhang Cao, Zeshuai Deng, Yanwu Xu, Mingkui Tan
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Deep neural networks have exhibited promising performance in image super-resolution (SR) by learning a nonlinear mapping function from low-resolution (LR) images to high-resolution (HR) images.  ...  Second, the paired LR-HR data may be unavailable in real-world applications and the underlying degradation method is often unknown.  ...  and unpaired real-world data demonstrate the effectiveness of the proposed dual regression scheme in image super-resolution.  ... 
doi:10.1109/cvpr42600.2020.00545 dblp:conf/cvpr/GuoCWCCDXT20 fatcat:7pavob63kndtxetrggnh6xr7um

Learning Omni-frequency Region-adaptive Representations for Real Image Super-Resolution [article]

Xin Li, Xin Jin, Tao Yu, Yingxue Pang, Simeng Sun, Zhizheng Zhang, Zhibo Chen
2021 arXiv   pre-print
Traditional single image super-resolution (SISR) methods that focus on solving single and uniform degradation (i.e., bicubic down-sampling), typically suffer from poor performance when applied into real-world  ...  The key to solving this more challenging real image super-resolution (RealSR) problem lies in learning feature representations that are both informative and content-aware.  ...  With the development of deep learning, a series of studies (Pan et al. 2020; Dong et al. 2015; Lim et al. 2017; Dong, Loy, and Tang 2016) have achieved great progress on traditional super-resolution  ... 
arXiv:2012.06131v2 fatcat:m7j6bu75lvek3ppj2gqrpjuu3u

Deep Degradation Prior for Real-World Super-Resolution

Kyungdeuk Ko, Bokyeung Lee, Jonghwan Hong, David K. Han, Hanseok Ko
2021 British Machine Vision Conference  
Real-world Super-Resolution (SR) is a very challenging task to reconstruct a higher resolution image from a real-world image which generally has unexpected artifacts and distortions.  ...  The methods estimate the noises and the blur kernels from real-world images to generate a new training set. However, these methods use the degradation only for dataset construction.  ...  Real-World Super-Resolution Real-world SR is a task that makes a blurry and noisy image taken with a real camera or smartphone into a clean HR image. Therefore, there is no LR-HR paired dataset.  ... 
dblp:conf/bmvc/KoLHHK21 fatcat:xv5kkflruzgy7amuslyor4x4xm

AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results [article]

Pengxu Wei, Hannan Lu, Radu Timofte, Liang Lin, Wangmeng Zuo, Zhihong Pan, Baopu Li, Teng Xi, Yanwen Fan, Gang Zhang, Jingtuo Liu, Junyu Han (+64 others)
2020 arXiv   pre-print
The goal is to attract more attention to realistic image degradation for the SR task, which is much more complicated and challenging, and contributes to real-world image super-resolution applications.  ...  They gauge the state-of-the-art approaches for real image SR in terms of PSNR and SSIM.  ...  ALONG The ALONG team proposed Dual Path Network with high frequency guided for real-world image Super-Resolution.  ... 
arXiv:2009.12072v1 fatcat:7cwsjfhqa5cf7avfvdabpmxrda

Learning a Degradation-Adaptive Network for Light Field Image Super-Resolution [article]

Yingqian Wang, Zhengyu Liang, Longguang Wang, Jungang Yang, Wei An, Yulan Guo
2022 arXiv   pre-print
Recent years have witnessed the great advances of deep neural networks (DNNs) in light field (LF) image super-resolution (SR).  ...  Extensive experiments on both synthetically degraded and real-world LFs demonstrate the effectiveness of our method.  ...  degradation models to train deep networks for real-world SR.  ... 
arXiv:2206.06214v1 fatcat:dlulqxvtxzhy3lhlnjpvazrq34

A Deep Residual Star Generative Adversarial Network for multi-domain Image Super-Resolution [article]

Rao Muhammad Umer, Asad Munir, Christian Micheloni
2021 arXiv   pre-print
To handle the multiple degradation, i.e. refers to multi-domain image super-resolution, we propose a deep Super-Resolution Residual StarGAN (SR2*GAN), a novel and scalable approach that super-resolves  ...  However, in real-world settings, the LR degradation process is unknown which can be bicubic LR, bilinear LR, nearest-neighbor LR, or real LR.  ...  In the real-world settings, the input LR image contains more complex degradation.  ... 
arXiv:2107.03145v1 fatcat:r6ql5ofkijevpcg2jgksjpzw5i

NTIRE 2019 Challenge on Real Image Super-Resolution: Methods and Results

Jianrui Cai, Shuhang Gu, Radu Timofte, Lei Zhang, Xiao Liu, Yukang Ding, Dongliang He, Chao Li, Yi Fu, Shilei Wen, Ruicheng Feng, Jinjin Gu (+92 others)
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
They gauge the state-of-the-art in real-world single image super-resolution.  ...  The challenge had 1 track, which was aimed at the real-world single image super-resolution problem with an unknown scaling factor.  ...  ., ETH Zurich, and The Hong Kong Polytechnic University. A. Teams and affiliations NTIRE2019 team  ... 
doi:10.1109/cvprw.2019.00274 dblp:conf/cvpr/CaiGTZa19 fatcat:tynb7gtpdvbfnbmugtelgcxude

Exploiting Style and Attention in Real-World Super-Resolution [article]

Xin Ma, Yi Li, Huaibo Huang, Mandi Luo, Ran He
2020 arXiv   pre-print
Real-world image super-resolution (SR) is a challenging image translation problem.  ...  To get real-world-like low-quality images paired with the HR images, we design the styleVAE to transfer the complex nuisance factors in real-world LR images to the generated LR images.  ...  In addition to styleVAE, we build a SR network for realworld super-resolution.  ... 
arXiv:1912.10227v2 fatcat:e5c5dbbmtjdfjdtdabsw4rrybm
« Previous Showing results 1 — 15 out of 10,412 results