Deep video deblurring for hand-held cameras

Conference Paper

Motion blur from camera shake is a major problem in videos captured by hand-held devices. Unlike single-image deblurring, video-based approaches can take advantage of the abundant information that exists across neighboring frames. As a result the best performing methods rely on the alignment of nearby frames. However, aligning images is a computationally expensive and fragile procedure, and methods that aggregate information must therefore be able to identify which regions have been accurately aligned and which have not, a task that requires high level scene understanding. In this work, we introduce a deep learning solution to video deblurring, where a CNN is trained end-toend to learn how to accumulate information across frames. To train this network, we collected a dataset of real videos recorded with a high frame rate camera, which we use to generate synthetic motion blur for supervision. We show that the features learned from this dataset extend to deblurring motion blur that arises due to camera shake in a wide range of videos, and compare the quality of results to a number of other baselines.

Full Text

Duke Authors

Cited Authors

  • Su, S; Delbracio, M; Wang, J; Sapiro, G; Heidrich, W; Wang, O

Published Date

  • November 6, 2017

Published In

  • Proceedings 30th Ieee Conference on Computer Vision and Pattern Recognition, Cvpr 2017

Volume / Issue

  • 2017-January /

Start / End Page

  • 237 - 246

International Standard Book Number 13 (ISBN-13)

  • 9781538604571

Digital Object Identifier (DOI)

  • 10.1109/CVPR.2017.33

Citation Source

  • Scopus