Papers › Local-Global Fusion Network for Video Super-Resolution

Local-Global Fusion Network for Video Super-Resolution

22 Sep 2020IEEE Access 2020 9archive 2025-07-28

Dewei Su, Hua Wang, Longcun Jin, Xianfang Sun, Xinyi Peng

The goal of video super-resolution technique is to address the problem of effectively restoring high-resolution (HR) videos from low-resolution (LR) ones. Previous methods commonly used optical flow to perform frame alignment and designed a framework from the perspective of space and time. However, inaccurate optical flow estimation may occur easily which leads to inferior restoration effects. In addition, how to effectively fuse the features of various video frames remains a challenging problem. In this paper, we propose a Local-Global Fusion Network (LGFN) to solve the above issues from a novel viewpoint. As an alternative to optical flow, deformable convolutions (DCs) with decreased multi-dilation convolution units (DMDCUs) are applied for efficient implicit alignment. Moreover, a structure with two branches, consisting of a Local Fusion Module (LFM) and a Global Fusion Module (GFM), is proposed to combine information from two different aspects. Specifically, LFM focuses on the relationship between adjacent frames and maintains the temporal consistency while GFM attempts to take advantage of all related features globally with a video shuffle strategy. Benefiting from our advanced network, experimental results on several datasets demonstrate that our LGFN can not only achieve comparative performance with state-of-the-art methods but also possess reliable ability on restoring a variety of video frames. The results on benchmark datasets of our LGFN are presented on https://github.com/BIOINSu/LGFN and the source code will be released as soon as the paper is accepted.

PaperPDFCode

Code

BIOINSu/LGFN mentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Optical Flow EstimationSuper-ResolutionVideo Super-Resolution

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over ERQA 18.342 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over LPIPS 11.759 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over MS-SSIM 0.889 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over PSNR 5.768 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over Subjective Score 2.944 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + vvenc BSQ-rate over VMAF 1.626 #44 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x264 BSQ-rate over ERQA 1.704 #49 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x264 BSQ-rate over LPIPS 1.324 #49 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x264 BSQ-rate over MS-SSIM 0.77 #49 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x264 BSQ-rate over PSNR 1.151 #49 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x264 BSQ-rate over VMAF 0.744 #49 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + uavs3e BSQ-rate over ERQA 9.279 #58 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + uavs3e BSQ-rate over LPIPS 4.504 #58 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + uavs3e BSQ-rate over MS-SSIM 2.427 #58 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + uavs3e BSQ-rate over PSNR 5.503 #58 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + uavs3e BSQ-rate over VMAF 1.625 #58 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x265 BSQ-rate over ERQA 13.213 #67 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x265 BSQ-rate over LPIPS 11.399 #67 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x265 BSQ-rate over MS-SSIM 1.533 #67 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x265 BSQ-rate over PSNR 6.646 #67 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + x265 BSQ-rate over VMAF 1.341 #67 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + aomenc BSQ-rate over ERQA 14.631 #72 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + aomenc BSQ-rate over LPIPS 5.536 #72 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + aomenc BSQ-rate over MS-SSIM 4.321 #72 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + aomenc BSQ-rate over PSNR 9.79 #72 of 85 Archive leaderboard report
Video Super-Resolution MSU Super-Resolution for Video Compression LGFN + aomenc BSQ-rate over VMAF 1.99 #72 of 85 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN 1 - LPIPS 0.903 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN ERQAv1.0 0.74 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN FPS 0.667 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN PSNR 31.291 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN QRCRv1.0 0.629 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN SSIM 0.898 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Super Resolution Benchmark: Detail Restoration LGFN Subjective score 6.505 #6 of 32 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement LGFN PSNR 27.42 #48 of 48 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement LGFN SSIM 0.939 #48 of 48 Archive leaderboard report
Video Super-Resolution MSU Video Upscalers: Quality Enhancement LGFN VMAF 57.79 #48 of 48 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Convolution

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections