Filters








2,270 Hits in 6.5 sec

Efficient Video Super-Resolution via Hierarchical Temporal Residual Networks

Zhi-Song Liu, Wan-Chi Siu, Yui-Lam Chan
2021 IEEE Access  
Super-Resolving (SR) video is more challenging compared with image super-resolution because of the demanding computation time.  ...  This paper proposes a lighter residual network, based on a multi-stage back projection for multi-frame SR.  ...  The objective of video super-resolution is to obtain a high-resolution clear video.  ... 
doi:10.1109/access.2021.3098326 fatcat:agpsehlckbdf7jpzvjuvsbmgaq

Ordinal Regression Based Subpixel Shift Estimation for Video Super-Resolution

Mithun Das Gupta, Shyamsundar Rajaram, Thomas S. Huang, Nemanja Petrovic
2007 EURASIP Journal on Advances in Signal Processing  
Finally, we demonstrate the applicability of our approach on superresolving synthetically generated images with global subpixel shifts and enhancing real video frames by accounting for both local integer  ...  We present a supervised learning-based approach for subpixel motion estimation which is then used to perform video superresolution.  ...  Applications, such as video frame registration, resolution enhancement, super-resolution, and optical-flow-based tracking, depend on reliable techniques for shift estimation for accuracy.  ... 
doi:10.1155/2007/85963 fatcat:76ot67iqxbh27i4tsdvpk4pi3q

MultiBoot Vsr: Multi-Stage Multi-Reference Bootstrapping for Video Super-Resolution

Ratheesh Kalarot, Fatih Porikli
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
super-resolution.  ...  Our method is ranked as the second best based on the SSIM scores on the NTIRE2019 VSR Challenge, Clean Track.  ...  Deep learning based methods build models on mainly external references while approximating complex nonlinear functions inherent in super-resolution task.  ... 
doi:10.1109/cvprw.2019.00258 dblp:conf/cvpr/KalarotP19 fatcat:44mzymchgzdxlo53s23tlczogq

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network [article]

Wenzhe Shi, Jose Caballero, Ferenc Huszár, Johannes Totz, Andrew P. Aitken, Rob Bishop, Daniel Rueckert, Zehan Wang
2016 arXiv   pre-print
Recently, several models based on deep neural networks have achieved great success in terms of both reconstruction accuracy and computational performance for single image super-resolution.  ...  We evaluate the proposed approach using images and videos from publicly available datasets and show that it performs significantly better (+0.15dB on Images and +0.39dB on Videos) and is an order of magnitude  ...  These methods, employ the back-propagation algorithm [22] to train on large image databases such as ImageNet [30] in order to learn nonlinear mappings of LR and HR image patches.  ... 
arXiv:1609.05158v2 fatcat:qxukr2hapfb7zg2b3njy6njudy

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

Wenzhe Shi, Jose Caballero, Ferenc Huszar, Johannes Totz, Andrew P. Aitken, Rob Bishop, Daniel Rueckert, Zehan Wang
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Recently, several models based on deep neural networks have achieved great success in terms of both reconstruction accuracy and computational performance for single image super-resolution.  ...  We evaluate the proposed approach using images and videos from publicly available datasets and show that it performs significantly better (+0.15dB on Images and +0.39dB on Videos) and is an order of magnitude  ...  These methods, employ the back-propagation algorithm [22] to train on large image databases such as ImageNet [30] in order to learn nonlinear mappings of LR and HR image patches.  ... 
doi:10.1109/cvpr.2016.207 dblp:conf/cvpr/ShiCHTABRW16 fatcat:uyftyo55bzginobslppgvp7p3a

Super-Resolution for "Jilin-1" Satellite Video Imagery via a Convolutional Network

Aoran Xiao, Zhongyuan Wang, Lei Wang, Yexian Ren
2018 Sensors  
Super-resolution for satellite video attaches much significance to earth observation accuracy, and the special imaging and transmission conditions on the video satellite pose great challenges to this task  ...  Meanwhile, we formulate a joint loss function by combining the output and high-dimensional features of a non-linear mapping network to precisely learn the desirable mapping relationship between low-resolution  ...  Instead, SSIM measures three aspects of image similarity, including brightness, contrast, and structure. Unlike PSNR, it is based on structural similarity rather than error sensitivity.  ... 
doi:10.3390/s18041194 pmid:29652838 pmcid:PMC5948634 fatcat:kunnegmdk5ejldk2wnls7sgneq

A Survey on Various Single Image Super Resolution Techniques
ENGLISH

A.Haza rathaiah
2013 International Journal of Innovative Research in Science, Engineering and Technology  
Super resolution methods which is generate high-resolution (HR) image from one or more low resolution images and various image quality metrics reviewed as measure the original image and reconstructed image  ...  In this paper, we presented different existing super resolution methods, positive and negative aspects of those methods, relevant work and methods of super resolution reconstruction method.  ...  A first order approximation of nonlinear mapping function,learned using local self-similar example patches is applied to low resolution image pathes for obtain the high resolution image patches.  ... 
doi:10.15680/ijirset.2012.0102024 fatcat:t45xr2uapvcrdnzds7ldc37eta

Generalized Face Super-Resolution

Kui Jia, Shaogang Gong
2008 IEEE Transactions on Image Processing  
Existing learning-based face super-resolution (hallucination) techniques generate high-resolution images of a single facial modality (i.e., at a fixed expression, pose and illumination) given one or set  ...  In particular, we formulate a unified tensor which can be reduced to two parts: a global image-based tensor for modeling the mappings among different facial modalities, and a local patch-based multiresolution  ...  Based on , generalized face super-resolution is patch-wisely performed in one single step. That is, face hallucination is a composition of its corresponding hallucinated local patches.  ... 
doi:10.1109/tip.2008.922421 pmid:18482883 fatcat:p4mjwtmtgngrjf23a27z6qt44i

Learning Hierarchical Decision Trees for Single-Image Super-Resolution

Jun-Jie Huang, Wan-Chi Siu
2017 IEEE transactions on circuits and systems for video technology (Print)  
Sparse representation has been extensively studied for image super-resolution (SR), and it achieved great improvement.  ...  Our proposed SR using decision tree (SRDT) method takes the divide-and-conquer strategy, which performs a few simple binary tests to classify an input low-resolution (LR) patch into one of the leaf nodes  ...  [21] proposed an SR method based on neighbor embedding (NE), which assumes that the LR patches and their corresponding HR patches share a similar low-dimensional nonlinear manifold.  ... 
doi:10.1109/tcsvt.2015.2513661 fatcat:zgvowkbwrrfc3ksj3uzrx2zvw4

Joint Prior Learning for Visual Sensor Network Noisy Image Super-Resolution

Bo Yue, Shuang Wang, Xuefeng Liang, Licheng Jiao, Caijin Xu
2016 Sensors  
Unlike conventional methods that only focus on upscaling images, JPISR alternatively solves upscaling mapping and denoising in the E-step and M-step.  ...  However, the captured images/videos are often low resolution with noise. Such visual data cannot be directly delivered to the advanced visual analysis.  ...  Author Contributions: Bo Yue proposed the original algorithm and wrote this paper; Shuang Wang and Xuefeng Liang revisited the paper and supervised a whole process; Licheng Jiao and Caijin Xu gave some  ... 
doi:10.3390/s16030288 pmid:26927114 pmcid:PMC4813863 fatcat:5fx6qwhkljhfdmnuuwpkdcqq4y

Fuzzy-rule based approach for single frame super resolution

Pulak Purkait, Bhabatosh Chanda
2013 2013 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)  
Our approach is based on the fundamental idea that a low-resolution (LR) image patch could be generated from any of the many possible high-resolution (HR) image patches.  ...  We do so by collecting a large amount of the LR-HR natural image patch pairs from an existing database, grouping them into different clusters and then generating fuzzy rules to get an efficient mapping  ...  We assume that LR patch feature space and HR patch feature space lie on the same manifold and use similar kind of mechanism for learning nonlinear mapping between input and output space as suggested by  ... 
doi:10.1109/fuzz-ieee.2013.6622394 dblp:conf/fuzzIEEE/PurkaitC13 fatcat:tw6zj4hrnbc3nnxew2ceuvpjdu

Anchored neighborhood deep network for single-image super-resolution

Wuzhen Shi, Shaohui Liu, Feng Jiang, Debin Zhao, Zhihong Tian
2018 EURASIP Journal on Image and Video Processing  
In this paper, we establish the relationship between the traditional sparse-representation-based single-image super-resolution methods and the deep-learning-based ones and use transfer learning to make  ...  As a result, each neuron will work on the same types of image patches that have similar details, which makes the network more accurate to recover high-frequency details.  ...  Acknowledgements We would like to acknowledge all our team members, especially Min Gao and Xinwei Gao, for their constructive suggestions on deep-learning-based image restoration and image compression.  ... 
doi:10.1186/s13640-018-0269-7 fatcat:spvm2uk5mjbetiurebb54gp7fy

A Group Variational Transformation Neural Network for Fractional Interpolation of Video Coding [article]

Sifeng Xia, Wenhan Yang, Yueyu Hu, Siwei Ma, Jiaying Liu
2018 arXiv   pre-print
Compared with the interpolation method of High Efficiency Video Coding, our method achieves 1.9% bit saving on average and up to 5.6% bit saving under low-delay P configuration.  ...  Motion compensation is an important technology in video coding to remove the temporal redundancy between coded video frames.  ...  The network first uniformly extracts a shared feature map from the input integer-position sample and then a group of copied shared feature maps are transformed to samples at various sub-pixel positions  ... 
arXiv:1806.07008v1 fatcat:ufyktfzkxvgh3czcxqyr3omgmy

Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution [article]

Wei-Sheng Lai, Jia-Bin Huang, Narendra Ahuja, Ming-Hsuan Yang
2017 arXiv   pre-print
At each pyramid level, our model takes coarse-resolution feature maps as input, predicts the high-frequency residuals, and uses transposed convolutions for upsampling to the finer level.  ...  In this paper, we propose the Laplacian Pyramid Super-Resolution Network (LapSRN) to progressively reconstruct the sub-band residuals of high-resolution images.  ...  [7] propose a Super-Resolution Convolutional Neural Network (SRCNN) to learn a nonlinear LR-to-HR mapping.  ... 
arXiv:1704.03915v2 fatcat:n6lhowwtfzcipol5jy3logsgrq

Bidirectional Recurrent Convolutional Networks for Multi-Frame Super-Resolution

Yan Huang, Wei Wang, Liang Wang
2015 Neural Information Processing Systems  
Super resolving a low-resolution video is usually handled by either single-image super-resolution (SR) or multi-frame SR.  ...  Single-Image SR deals with each video frame independently, and ignores intrinsic temporal dependency of video frames which actually plays a very important role in video super-resolution.  ...  Acknowledgments This work is jointly supported by National Natural Science Foundation of China (61420106015, 61175003, 61202328, 61572504) and National Basic Research Program of China (2012CB316300).  ... 
dblp:conf/nips/HuangWW15 fatcat:alpoxuoawvfolgo4aqawl7rn2q
« Previous Showing results 1 — 15 out of 2,270 results