Filters








110 Hits in 6.1 sec

Cross-Scale Residual Network for Multiple Tasks:Image Super-resolution, Denoising, and Deblocking [article]

Yuan Zhou, Xiaoting Du, Yeda Zhang, Sun-Yuan Kung
2019 arXiv   pre-print
It is desirable for an image processing network to support well with three vital tasks, namely, super-resolution, denoising, and deblocking.  ...  Recently, deep convolutional neural networks have proven promising for such learning processing.  ...  [14] uses CNN-based denoisers into model-based optimization for image denoising and super-resolution.  ... 
arXiv:1911.01257v1 fatcat:nnctvxez4bfmjjgggiox3m6ooi

Video Enhancement with Task-Oriented Flow

Tianfan Xue, Baian Chen, Jiajun Wu, Donglai Wei, William T. Freeman
2019 International Journal of Computer Vision  
TOFlow outperforms traditional optical flow on standard benchmarks as well as our Vimeo-90K dataset in three video processing tasks: frame interpolation, video denoising/deblocking, and video super-resolution  ...  We design a neural network with a trainable motion estimation component and a video processing component, and train them jointly to learn the task-oriented flow.  ...  Acknowledgements This work is supported by NSF RI-1212849, NSF BIGDATA-1447476, Facebook, Shell Research, and Toyota Research Institute.  ... 
doi:10.1007/s11263-018-01144-2 fatcat:wyexetarjjhvzgfdr3pajfmn3a

LAPAR: Linearly-Assembled Pixel-Adaptive Regression Network for Single Image Super-Resolution and Beyond [article]

Wenbo Li, Kun Zhou, Lu Qi, Nianjuan Jiang, Jiangbo Lu, Jiaya Jia
2021 arXiv   pre-print
Moreover, based on the same idea, LAPAR is extended to tackle other restoration tasks, e.g., image denoising and JPEG image deblocking, and again, yields strong performance.  ...  Single image super-resolution (SISR) deals with a fundamental problem of upsampling a low-resolution (LR) image to its high-resolution (HR) version.  ...  Conclusion We have presented a linearly-assembled pixel-adaptive regression network (LAPAR) for image super-resolution.  ... 
arXiv:2105.10422v1 fatcat:erwwgoyaajchjjqzmn7uwoa65y

EDPN: Enhanced Deep Pyramid Network for Blurry Image Restoration [article]

Ruikang Xu, Zeyu Xiao, Jie Huang, Yueyi Zhang, Zhiwei Xiong
2021 arXiv   pre-print
Experimental results demonstrate that our method significantly outperforms existing solutions for blurry image super-resolution and blurry image deblocking.  ...  To address these challenges, we propose an Enhanced Deep Pyramid Network (EDPN) for blurry image restoration from multiple degradations, by fully exploiting the self- and cross-scale similarities in the  ...  Acknowledgement We acknowledge funding from National Key R&D Program of China under Grant 2017YFA0700800, and National Natural Science Foundation of China under Grant 61901435.  ... 
arXiv:2105.04872v1 fatcat:4jjp7s5pjbd5flw7ar3sed46qu

Compression Artifacts Removal Using Convolutional Neural Networks [article]

Pavel Svoboda and Michal Hradis and David Barina and Pavel Zemcik
2016 arXiv   pre-print
This paper shows that it is possible to train large and deep convolutional neural networks (CNN) for JPEG compression artifacts reduction, and that such networks can provide significantly better reconstruction  ...  quality compared to previously used smaller networks as well as to any other state-of-the-art methods.  ...  TE01020415) and the Ministry of Education, Youth and Sports from the National Programme of Sustainability (NPU II) project IT4Innovations excellence in science (no. LQ1602).  ... 
arXiv:1605.00366v1 fatcat:fjsdzrl2rvafrhkzg7273b3lcu

Edge-Aware Image Compression using Deep Learning-based Super-resolution Network [article]

Dipti Mishra, Satish Kumar Singh, Rajat Kumar Singh, Krishna Preetham
2021 arXiv   pre-print
in prior works & (b) a super-resolution convolutional neural network (CNN) for post-processing along with a corresponding pre-processing network for improved rate-distortion performance in the low rate  ...  We propose a learning-based compression scheme that envelopes a standard codec between pre and post-processing deep CNNs.  ...  Different from prior works, however, we: In the proposed approach, we utilize the enhanced deep super-resolution (EDSR) network [29] , which is a multi-scale network designed for specific super-resolution  ... 
arXiv:2104.04926v1 fatcat:j4zwhcl5mfbyra4boilphlu5ym

Deep Convolution Networks for Compression Artifacts Reduction [article]

Ke Yu, Chao Dong, Chen Change Loy, Xiaoou Tang
2016 arXiv   pre-print
Inspired by the success of deep convolutional networks (DCN) on superresolution, we formulate a compact and efficient network for seamless attenuation of different compression artifacts.  ...  Our method shows superior performance than the state-of-the-art methods both on benchmark datasets and a real-world use case.  ...  As a recent method for image super-resolution [34] , A+ [29] has also been successfully applied for compression artifacts reduction.  ... 
arXiv:1608.02778v1 fatcat:okb4eqezabcgzif3yxbszwdq74

MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement [article]

Wenbo Bao, Wei-Sheng Lai, Xiaoyun Zhang, Zhiyong Gao, Ming-Hsuan Yang
2019 arXiv   pre-print
., super-resolution, denoising, and deblocking.  ...  Recently, a number of data-driven frame interpolation methods based on convolutional neural networks have been proposed.  ...  The DeepSR [51] and ToFlow [9] methods are CNN-based approaches for video super-resolution.  ... 
arXiv:1810.08768v2 fatcat:yhsvq7qb7jgznjthfsaxxyxtze

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

Kai Zhang, Wangmeng Zuo, Yunjin Chen, Deyu Meng, Lei Zhang
2017 IEEE Transactions on Image Processing  
This property motivates us to train a single DnCNN model to tackle with several general image denoising tasks such as Gaussian denoising, single image super-resolution and JPEG image deblocking.  ...  , and regularization method into image denoising.  ...  Among the above deep neural networks based methods, MLP and TNRD can achieve promising performance and are able to compete with BM3D.  ... 
doi:10.1109/tip.2017.2662206 pmid:28166495 fatcat:7opuakoakndnla5lre35yu72si

DPW-SDNet: Dual Pixel-Wavelet Domain Deep CNNs for Soft Decoding of JPEG-Compressed Images

Honggang Chen, Xiaohai He, Linbo Qing, Shuhua Xiong, Truong Q. Nguyen
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
soft decoding network for JPEGcompressed images, namely DPW-SDNet.  ...  Inspired by the excellent performance of the deep convolutional neural networks (CNNs) on both low-level and high-level computer vision problems, we develop a dual pixel-wavelet domain deep CNNs-based  ...  [17] introduced the residual learning technique and designed a very deep network of twenty layers for single image super-resolution.  ... 
doi:10.1109/cvprw.2018.00114 dblp:conf/cvpr/ChenHQXN18 fatcat:rllwqhn2hbdmzckqh5biq7qyxm

CISRDCNN: Super-resolution of compressed images using deep convolutional neural networks [article]

Honggang Chen, Xiaohai He, Chao Ren, Linbo Qing, Qizhi Teng
2017 arXiv   pre-print
In this paper, an end-to-end trainable deep convolutional neural network is designed to perform SR on compressed images (CISRDCNN), which reduces compression artifacts and improves image resolution jointly  ...  In recent years, much research has been conducted on image super-resolution (SR). To the best of our knowledge, however, few SR methods were concerned with compressed images.  ...  To the best of our knowledge, nevertheless, very few research has been done on deep neural networks-based SR methods for compressed images.  ... 
arXiv:1709.06229v1 fatcat:ea6b3t26jfdnjaae6stm6r677q

DPW-SDNet: Dual Pixel-Wavelet Domain Deep CNNs for Soft Decoding of JPEG-Compressed Images [article]

Honggang Chen and Xiaohai He and Linbo Qing and Shuhua Xiong and Truong Q. Nguyen
2018 arXiv   pre-print
soft decoding network for JPEG-compressed images, namely DPW-SDNet.  ...  Inspired by the excellent performance of the deep convolutional neural networks (CNNs) on both low-level and high-level computer vision problems, we develop a dual pixel-wavelet domain deep CNNs-based  ...  [17] introduced the residual learning technique and designed a very deep network of twenty layers for single image super-resolution.  ... 
arXiv:1805.10558v1 fatcat:sitqnybqmbeexpr36yuoe2yyua

An End-to-End Compression Framework Based on Convolutional Neural Networks [article]

Feng Jiang, Wen Tao, Shaohui Liu, Jie Ren, Xun Guo, Debin Zhao
2017 arXiv   pre-print
Deep learning, e.g., convolutional neural networks (CNNs), has achieved great success in image processing and computer vision especially in high level vision applications such as recognition and understanding  ...  post-processing methods.  ...  Image Super-Resolution Based on Deep Learning Recently, CNNs have been used successfully for image super-resolution (SR) especially when residual learning [18] and gradients-based optimization algorithms  ... 
arXiv:1708.00838v1 fatcat:xr6yssotmrbxzm47lyebdhfvye

Trainable Nonlinear Reaction Diffusion: A Flexible Framework for Fast and Effective Image Restoration

Yunjin Chen, Thomas Pock
2017 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We demonstrate its capabilities with three representative applications, Gaussian image denoising, single image super resolution and JPEG deblocking.  ...  We describe a flexible learning framework based on the concept of nonlinear reaction diffusion models for various image restoration problems.  ...  We have trained our models for the problem of Gaussian denoising, single image super resolution and JPEG deblocking.  ... 
doi:10.1109/tpami.2016.2596743 pmid:27529868 fatcat:o6loggo7inhb5g4ffwyv3g5vsi

Decouple Learning for Parameterized Image Operators [article]

Qingnan Fan, Dongdong Chen, Lu Yuan, Gang Hua, Nenghai Yu, Baoquan Chen
2018 arXiv   pre-print
Many different deep networks have been used to approximate, accelerate or improve traditional image operators, such as image smoothing, super-resolution and denoising.  ...  To overcome this limitation, we propose a new decouple learning algorithm to learn from the operator parameters to dynamically adjust the weights of a deep network for image operators, denoted as the base  ...  In this paper we deal with four representative tasks in this venue: super resolution [14, 27] , denoising [26, 33] , deblocking [13, 38] and derain [20, 49] , which have been studied with deep learning  ... 
arXiv:1807.08186v2 fatcat:rrhy3uvmfjdkhnokhakwsmds34
« Previous Showing results 1 — 15 out of 110 results