Skip to main navigation Skip to search Skip to main content

Spatio-Temporal Deformable Attention Network for Video Deblurring

  • Huicong Zhang
  • , Haozhe Xie
  • , Hongxun Yao*
  • *Corresponding author for this work
  • Harbin Institute of Technology
  • Tencent

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The key success factor of the video deblurring methods is to compensate for the blurry pixels of the mid-frame with the sharp pixels of the adjacent video frames. Therefore, mainstream methods align the adjacent frames based on the estimated optical flows and fuse the alignment frames for restoration. However, these methods sometimes generate unsatisfactory results because they rarely consider the blur levels of pixels, which may introduce blurry pixels from video frames. Actually, not all the pixels in the video frames are sharp and beneficial for deblurring. To address this problem, we propose the spatio-temporal deformable attention network (STDANet) for video delurring, which extracts the information of sharp pixels by considering the pixel-wise blur levels of the video frames. Specifically, STDANet is an encoder-decoder network combined with the motion estimator and spatio-temporal deformable attention (STDA) module, where motion estimator predicts coarse optical flows that are used as base offsets to find the corresponding sharp pixels in STDA module. Experimental results indicate that the proposed STDANet performs favorably against state-of-the-art methods on the GoPro, DVD, and BSD datasets.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2022 - 17th European Conference, Proceedings
EditorsShai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, Tal Hassner
PublisherSpringer Science and Business Media Deutschland GmbH
Pages581-596
Number of pages16
ISBN (Print)9783031197864
DOIs
StatePublished - 2022
Event17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, Israel
Duration: 23 Oct 202227 Oct 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13676 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th European Conference on Computer Vision, ECCV 2022
Country/TerritoryIsrael
CityTel Aviv
Period23/10/2227/10/22

Keywords

  • Pixel-wise blur levels
  • Spatio-temporal deformable attention
  • Video deblurring

Fingerprint

Dive into the research topics of 'Spatio-Temporal Deformable Attention Network for Video Deblurring'. Together they form a unique fingerprint.

Cite this