Filters








1,493 Hits in 4.7 sec

Learning for Video Super-Resolution through HR Optical Flow Estimation [article]

Longguang Wang, Yulan Guo, Zaiping Lin, Xinpu Deng, Wei An
2018 arXiv   pre-print
However, LR optical flows are used in existing deep learning based methods for correspondence generation.  ...  Video super-resolution (SR) aims to generate a sequence of high-resolution (HR) frames with plausible and temporally consistent details from their low-resolution (LR) counterparts.  ...  Introduction Super-resolution (SR) aims to generate high-resolution (HR) images or videos from their low-resolution (LR) counterparts.  ... 
arXiv:1809.08573v2 fatcat:hlx4xzmwkzd35hyxdhjb7lb3eu

BFRVSR: A Bidirectional Frame Recurrent Method for Video Super-Resolution

Xiongxiong Xue, Zhenqi Han, Weiqin Tong, Mingqi Li, Lizhuang Liu
2020 Applied Sciences  
In this work, a bidirectional frame recurrent video super-resolution method is proposed.  ...  Video super-resolution is a challenging task. One possible solution, called the sliding window method, tries to divide the generation of high-resolution video sequences into independent subtasks.  ...  [6] proposed a recursive algorithm for video super-resolution. The FRVSR [6] network estimates the optical flow F LR t→t−1 of I LR t−1 and I LR t , and uses I HR t−1 .  ... 
doi:10.3390/app10238749 fatcat:bqgrkgkuqrhgpb3o6huplz4oeq

Video Super Resolution Based on Deep Learning: A Comprehensive Survey [article]

Hongying Liu, Zhubo Ruan, Peng Zhao, Chao Dong, Fanhua Shang, Yuanyuan Liu, Linlin Yang, Radu Timofte
2022 arXiv   pre-print
It is well known that the leverage of information within video frames is important for video super-resolution.  ...  In this survey, we comprehensively investigate 33 state-of-the-art video super-resolution (VSR) methods based on deep learning.  ...  Yaowei Wang (Associate Professor with Peng Cheng Laboratory, Shenzhen, China) for their help in improving the quality of this manuscript.  ... 
arXiv:2007.12928v3 fatcat:nxoejcfdnzas3jznbqsale36ty

Efficient Space-time Video Super Resolution using Low-Resolution Flow and Mask Upsampling [article]

Saikat Dutta, Nisarg A. Shah, Anurag Mittal
2021 arXiv   pre-print
This paper explores an efficient solution for Space-time Super-Resolution, aiming to generate High-resolution Slow-motion videos from Low Resolution and Low Frame rate videos.  ...  Input LR frames are super-resolved using a state-of-the-art Video Super-Resolution method.  ...  [39] learned self-supervised task-specific optical flow for various Video enhancement problems, including temporal interpolation. Niklaus et al.  ... 
arXiv:2104.05778v3 fatcat:cbgo3l5sxbdsdf77vcogmesx5y

Frame-Recurrent Video Super-Resolution [article]

Mehdi S. M. Sajjadi and Raviteja Vemulapalli and Matthew Brown
2018 arXiv   pre-print
In this work, we propose an end-to-end trainable frame-recurrent video super-resolution framework that uses the previously inferred HR estimate to super-resolve the subsequent frame.  ...  Recent advances in video super-resolution have shown that convolutional neural networks combined with motion compensation are able to merge information from multiple low-resolution (LR) frames to generate  ...  the same optical flow and super-resolution networks.  ... 
arXiv:1801.04590v4 fatcat:l3b2fk3cyjeyxok3rbyhzeieyu

Deep Video Super-Resolution using HR Optical Flow Estimation [article]

Longguang Wang, Yulan Guo, Li Liu, Zaiping Lin, Xinpu Deng, Wei An
2020 arXiv   pre-print
Existing deep learning based methods commonly estimate optical flows between LR frames to provide temporal dependency.  ...  Video super-resolution (SR) aims at generating a sequence of high-resolution (HR) frames with plausible and temporally consistent details from their low-resolution (LR) counterparts.  ...  Flow Estimation.: To obtain HR optical flows from LR inputs, an alternative is to perform single image super-resolution (SISR) on separated LR frames first and then estimate HR optical flows from these  ... 
arXiv:2001.02129v1 fatcat:a5n3elabhbhf3l7emwayfc3oh4

Medical Video Super-Resolution Based on Asymmetric Back-Projection Network with Multilevel Error Feedback

Sheng Ren, Jianqi Li, Kehua Guo, Fangfang Li
2021 IEEE Access  
We construct a single-frame medical video super-resolution model as the benchmark model, combine the optical flow algorithm and multiframe fusion strategy to propose a medical video super-resolution method  ...  Medical video is important for medical diagnosis.  ...  This article is extended from the Conference paper written by Sheng Ren in The 12th IEEE International Conference on Cyber, Physical and Social Computing (Towards Efficient Medical Video Super-Resolution  ... 
doi:10.1109/access.2021.3054433 fatcat:fyju5efvdbe5fd4q35i7m3d7vq

Frame and Feature-Context Video Super-Resolution [article]

Bo Yan, Chuming Lin, Weimin Tan
2019 arXiv   pre-print
For video super-resolution, current state-of-the-art approaches either process multiple low-resolution (LR) frames to produce each output high-resolution (HR) frame separately in a sliding window fashion  ...  or recurrently exploit the previously estimated HR frames to super-resolve the following frame.  ...  The goal in image and video super-resolution (SR) is to reconstruct a high-resolution (HR) image or video from its down-sampled low-resolution (LR) version.  ... 
arXiv:1909.13057v1 fatcat:yqqyeeukojevxh7mt6nmgscavq

Frame and Feature-Context Video Super-Resolution

Bo Yan, Chuming Lin, Weimin Tan
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
For video super-resolution, current state-of-the-art approaches either process multiple low-resolution (LR) frames to produce each output high-resolution (HR) frame separately in a sliding window fashion  ...  or recurrently exploit the previously estimated HR frames to super-resolve the following frame.  ...  The goal in image and video super-resolution (SR) is to reconstruct a high-resolution (HR) image or video from its down-sampled low-resolution (LR) version.  ... 
doi:10.1609/aaai.v33i01.33015597 fatcat:l2cikopu6bcuzi3ocetrmtx6ke

Super resolution reconstruction algorithm of UAV image based on residual neural network

Xiaorong Zhang, Yueqi Ma
2021 IEEE Access  
with the predicted high-resolution optical flow.  ...  In low resolution space, the dense residual network is used to obtain the complementary information of adjacent video frames, and then the optical flow of highresolution video frames is predicted through  ...  super-resolution fusion network for better fusion of video frames, and the loss function of the network and the super-resolution fusion module is estimated by using the sum of optical flow residuals.  ... 
doi:10.1109/access.2021.3114437 fatcat:7zvontrovrgtlkqpepb3hfqhbi

DSTnet: Deformable Spatio-Temporal Convolutional Residual Network for Video Super-Resolution

Anusha Khan, Allah Bux Sargano, Zulfiqar Habib
2021 Mathematics  
Video super-resolution (VSR) aims at generating high-resolution (HR) video frames with plausible and temporally consistent details using their low-resolution (LR) counterparts, and neighboring frames.  ...  These methods cannot fully exploit the spatio-temporal information that significantly affects the quality of resultant HR videos.  ...  Earlier VSR methods estimated motion through separate multiple optical flow algorithms [5] and enabled an end-to-end trainable process for VSR.  ... 
doi:10.3390/math9222873 fatcat:dlh2t4fs25bbvp55wb3mbgt42y

Prediction-assistant Frame Super-Resolution for Video Streaming [article]

Wang Shen, Wenbo Bao, Guangtao Zhai, Charlie L Wang, Jerry W Hu, Zhiyong Gao
2021 arXiv   pre-print
., lossy), we propose to use previously received high-resolution frames to enhance the low-quality current ones. For the first case, we propose a small yet effective video frame prediction network.  ...  For the second case, we improve the video prediction network to a video enhancement network to associate current frames as well as previous frames to restore high-quality images.  ...  Moreover, we learn the residual optical flow using the U-net, which can be added to the initial estimated optical flow for refinement.  ... 
arXiv:2103.09455v1 fatcat:53if7lovwnaxplhtos54vsh4lq

Deep Gradient Prior Regularized Robust Video Super-Resolution

Qiang Song, Hangfan Liu
2021 Electronics  
This paper proposes a robust multi-frame video super-resolution (SR) scheme to obtain high SR performance under large upscaling factors.  ...  A forward and backward motion field prior is used to regularize the estimation of the motion flow between frames.  ...  Figure 5 . 5 Optical flow estimation results (color coded) of video "city".  ... 
doi:10.3390/electronics10141641 fatcat:xp2igllfr5ftbbrbwzswt3w6oi

Deformable Non-local Network For Video Super-Resolution [article]

Hua Wang, Dewei Su, Longcun Jin, Chuangchuang Liu
2019 arXiv   pre-print
The video super-resolution (VSR) task aims to restore a high-resolution video frame by using its corresponding low-resolution frame and multiple neighboring frames.  ...  At present, many deep learning-based VSR methods rely on optical flow to perform frame alignment. The final recovery results will be greatly affected by the accuracy of optical flow.  ...  It is a non-flow-based method for effective and efficient video super-resolutions.  ... 
arXiv:1909.10692v1 fatcat:kvmdpw3gpjhyvitfj7cjhc7e4e

HSTR-Net: High Spatio-Temporal Resolution Video Generation For Wide Area Surveillance [article]

H. Umut Suluhan, Hasan F. Ates, Bahadir K. Gunturk
2022 arXiv   pre-print
In this paper we propose an end-to-end trainable deep network that performs optical flow estimation and frame reconstruction by combining inputs from both video feeds.  ...  This paper presents the usage of multiple video feeds for the generation of HSTR video as an extension of reference based super resolution (RefSR).  ...  Applying single image super-resolution (SISR) directly to each LR video frame produces HR video but lacks temporal coherency.  ... 
arXiv:2204.04435v1 fatcat:wxjnbtnccratjiv6hqn2dksbuy
« Previous Showing results 1 — 15 out of 1,493 results