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

Video deblurring models exploit information in the neighboring frames toremove blur caused by the motion of the camera and the objects. RecurrentNeural Networks~(RNNs) are often adopted to model the temporal dependencybetween frames via hidden states. When motion blur is strong, however, hiddenstates are hard to deliver proper information due to the displacement betweendifferent frames. While there have been attempts to update the hidden states,it is difficult to handle misaligned features beyond the receptive field ofsimple modules. Thus, we propose 2 modules to supplement the RNN architecturefor video deblurring. First, we design Ping-Pong RNN~(PPRNN) that acts onupdating the hidden states by referring to the features from the current andthe previous time steps alternately. PPRNN gathers relevant information fromthe both features in an iterative and balanced manner by utilizing itsrecurrent architecture. Second, we use a Selective Non-Local Attention~(SNLA)module to additionally refine the hidden state by aligning it with thepositional information from the input frame feature. The attention score isscaled by the relevance to the input feature to focus on the necessaryinformation. By paying attention to hidden states with both modules, which havestrong synergy, our PAHS framework improves the representation powers of RNNstructures and achieves state-of-the-art deblurring performance on standardbenchmarks and real-world videos.

Pay Attention to Hidden States for Video Deblurring: Ping-Pong Recurrent Neural Networks and Selective Non-Local Attention