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

State-of-the-art video deblurring methods cannot handle blurry videosrecorded in dynamic scenes, since they are built under a strong assumption thatthe captured scenes are static. Contrary to the existing methods, we propose avideo deblurring algorithm that can deal with general blurs inherent in dynamicscenes. To handle general and locally varying blurs caused by various sources,such as moving objects, camera shake, depth variation, and defocus, we estimatepixel-wise non-uniform blur kernels. We infer bidirectional optical flows tohandle motion blurs, and also estimate Gaussian blur maps to remove opticalblur from defocus in our new blur model. Therefore, we propose a single energymodel that jointly estimates optical flows, defocus blur maps and latentframes. We also provide a framework and efficient solvers to minimize theproposed energy model. By optimizing the energy model, we achieve significantimprovements in removing general blurs, estimating optical flows, and extendingdepth-of-field in blurry frames. Moreover, in this work, to evaluate theperformance of non-uniform deblurring methods objectively, we have constructeda new realistic dataset with ground truths. In addition, extensive experimentalon publicly available challenging video data demonstrate that the proposedmethod produces qualitatively superior performance than the state-of-the-artmethods which often fail in either deblurring or optical flow estimation.

Dynamic Scene Deblurring using a Locally Adaptive Linear Blur Model