English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We consider the challenging task of training models for image-to-videodeblurring, which aims to recover a sequence of sharp images corresponding to agiven blurry image input. A critical issue disturbing the training of animage-to-video model is the ambiguity of the frame ordering since both theforward and backward sequences are plausible solutions. This paper proposes aneffective self-supervised ordering scheme that allows training high-qualityimage-to-video deblurring models. Unlike previous methods that rely onorder-invariant losses, we assign an explicit order for each video sequence,thus avoiding the order-ambiguity issue. Specifically, we map each videosequence to a vector in a latent high-dimensional space so that there exists ahyperplane such that for every video sequence, the vectors extracted from itand its reversed sequence are on different sides of the hyperplane. The side ofthe vectors will be used to define the order of the corresponding sequence.Last but not least, we propose a real-image dataset for the image-to-videodeblurring problem that covers a variety of popular domains, including face,hand, and street. Extensive experimental results confirm the effectiveness ofour method. Code and data are available athttps://github.com/VinAIResearch/HyperCUT.git

HyperCUT: Video Sequence from a Single Blurry Image using Unsupervised Ordering