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

The previous deep video compression approaches only use the single scalemotion compensation strategy and rarely adopt the mode prediction techniquefrom the traditional standards like H.264/H.265 for both motion and residualcompression. In this work, we first propose a coarse-to-fine (C2F) deep videocompression framework for better motion compensation, in which we performmotion estimation, compression and compensation twice in a coarse to finemanner. Our C2F framework can achieve better motion compensation resultswithout significantly increasing bit costs. Observing hyperprior information(i.e., the mean and variance values) from the hyperprior networks containsdiscriminant statistical information of different patches, we also propose twoefficient hyperprior-guided mode prediction methods. Specifically, usinghyperprior information as the input, we propose two mode prediction networks torespectively predict the optimal block resolutions for better motion coding anddecide whether to skip residual information from each block for better residualcoding without introducing additional bit cost while bringing negligible extracomputation cost. Comprehensive experimental results demonstrate our proposedC2F video compression framework equipped with the new hyperprior-guided modeprediction methods achieves the state-of-the-art performance on HEVC, UVG andMCL-JCV datasets.

Coarse-to-fine Deep Video Coding with Hyperprior-guided Mode Prediction