CoChat
Sign Up
Cited by 1,272Advanced Image Processing TechniquesImage Enhancement TechniquesGenerative Adversarial Networks and Image Synthesis

The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.

EDVR: Video Restoration With Enhanced Deformable Convolutional Networks

Wang, Xintao, Chan, Kelvin C.K., Yu, Ke, Dong, Chao, Loy, Chen Change·2019·2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
DOI 10.1109/cvprw.2019.00247
Open this paper in CoChat — free account

Abstract

Video restoration tasks, including super-resolution, deblurring, etc, are drawing increasing attention in the computer vision community. A challenging benchmark named REDS is released in the NTIRE19 Challenge. This new benchmark challenges existing methods from two aspects: (1) how to align multiple frames given large motions, and (2) how to effectively fuse different frames with diverse motion and blur. In this work, we propose a novel Video Restoration framework with Enhanced Deformable convolutions, termed EDVR, to address these challenges. First, to handle large motions, we devise a Pyramid, Cascading and Deformable (PCD) alignment module, in which frame alignment is done at the feature level using deformable convolutions in a coarse-to-fine manner. Second, we propose a Temporal and Spatial Attention (TSA) fusion module, in which attention is applied both temporally and spatially, so as to emphasize important features for subsequent restoration. Thanks to these modules, our EDVR wins the champions and outperforms the second place by a large margin in all four tracks in the NTIRE19 video restoration and enhancement challenges. EDVR also demonstrates superior performance to state-of-the-art published methods on video super-resolution and deblurring. The code is available at https://github.com/xinntao/EDVR.

Overview

Integrity signals

Registry

Crossref · 10.1109/cvprw.2019.00247

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 48

  1. 10.1109/CVPR.2018.00853 · DOI
  2. 10.1109/CVPR.2018.00853 · DOI
  3. 10.1109/CVPR.2011.5995360 · DOI
  4. 10.1109/CVPR.2018.00693 · DOI
  5. ranjan, 2016, Optical flow estimation using a spatial pyramid network
  6. 10.1109/CVPR.2017.739 · DOI
  7. 10.1109/ICCV.2017.479 · DOI
  8. 10.1109/TIP.2009.2023703 · DOI
  9. 10.1109/CVPR.2018.00931 · DOI
  10. 10.1109/CVPR.2017.33 · DOI
  11. 10.1109/CVPRW.2019.00249 · DOI
  12. 10.1109/CVPR.2017.35 · DOI
  13. 10.1109/CVPRW.2019.00250 · DOI
  14. caballero, 0, Realtime video super-resolution with spatio-temporal networks and motion compensation, CVPR
  15. 10.1007/978-3-030-01258-8_21 · DOI
  16. 10.1109/CVPRW.2017.151 · DOI
  17. 10.1109/ICCV.2017.274 · DOI
  18. 10.1109/TPAMI.2013.127 · DOI
  19. ma, 0, Handling motion blur in multi-frame super-resolution, CVPR
  20. liu, 0, Non-local recurrent network for image restoration, NIPS
  21. 10.1109/CVPRW.2019.00251 · DOI
  22. molchanov, 2016, Pruning convolutional neural networks for resource efficient inference
  23. 10.1109/TIP.2018.2867733 · DOI
  24. zhang, 0, Image super-resolution using very deep residual channel attention networks, ECCV
  25. zhao, 0, Trajectory convolution for action recognition, NIPS
  26. 10.1109/CVPR.2018.00340 · DOI
  27. 10.1109/TCI.2016.2532323 · DOI
  28. tian, 2018, TDAN Temporally deformable alignment network for video super-resolution
  29. 10.1109/TPAMI.2017.2761348 · DOI
  30. kim, 0, Spatio-temporal transformer network for video restoration, ECCV
  31. kupyn, 2017, DeblurGAN blind motion deblurring using conditional adversarial networks
  32. 10.1109/CVPR.2018.00854 · DOI
  33. 10.1109/CVPR.2017.618 · DOI
  34. 10.1109/ICCV.2015.68 · DOI
  35. 10.1109/ICIP.2015.7350764 · DOI
  36. 10.1109/CVPR.2019.00402 · DOI
  37. dong, 0, Learning a deep convolutional network for image super-resolution, ECCV
  38. hui, 2019, A Lightweight Optical Flow CNN - Revisiting Data Fidelity and Regularization
  39. hui, 0, Lite-flownet: A lightweight convolutional neural network for optical flow estimation, CVPR
  40. 10.1109/CVPR.2018.00259 · DOI
  41. 10.1109/CVPR.2017.179 · DOI
  42. wang, 0, Esrgan: Enhanced super-resolution generative adversarial networks, ECCVW
  43. 10.1109/CVPR.2018.00070 · DOI
  44. xue, 2017, Video Enhancement with Task-Oriented Flow
  45. woo, 0, Cbam: Convolutional block attention module, ECCV
  46. 10.1109/CVPR.2016.206 · DOI
  47. 10.1109/CVPRW.2017.149 · DOI
  48. vaswani, 0, Attention is all you need, NIPS

Cited by

Sorted by citations

Metadata

Authors
Wang, X., Chan, K. C. K., Yu, K., Dong, C., & Loy, C. C.
Affiliations
  • Chinese University of Hong Kong (HK)
  • Nanyang Technological University (SG)
  • Chinese Academy of Sciences (CN)
ORCID
Venue
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Publisher
IEEE
Published
2019-06
Pages
1954-1963
DOI
10.1109/cvprw.2019.00247
OpenAlex
W2965669158
Cited by
1,272
References
5
Citations by year
2026: 96 · 2025: 183 · 2024: 230 · 2023: 228 · 2022: 205 · 2021: 215 · 2020: 100 · 2019: 15
FWCI
43.63
Type
conference paper
Language
en
Open access
closed
Licence
Locations
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image Processing Techniques
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Enhancement Techniques
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
Keywords
video restoration, video super-resolution, video deblurring, deformable convolution, frame alignment, coarse-to-fine alignment, frame fusion, large motion

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review