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

In-loop filtering is used in video coding to process the reconstructed framein order to remove blocking artifacts. With the development of convolutionalneural networks (CNNs), CNNs have been explored for in-loop filteringconsidering it can be treated as an image de-noising task. However, in additionto being a distorted image, the reconstructed frame is also obtained by a fixedline of block based encoding operations in video coding. It carries coding-unitbased coding distortion of some similar characteristics. Therefore, in thispaper, we address the filtering problem from two aspects, global appearancerestoration for disrupted texture and local coding distortion restorationcaused by fixed pipeline of coding. Accordingly, a three-stream globalappearance and local coding distortion based fusion network is developed with ahigh-level global feature stream, a high-level local feature stream and alow-level local feature stream. Ablation study is conducted to validate thenecessity of different features, demonstrating that the global features andlocal features can complement each other in filtering and achieve betterperformance when combined. To the best of our knowledge, we are the first onethat clearly characterizes the video filtering process from the above globalappearance and local coding distortion restoration aspects with experimentalverification, providing a clear pathway to developing filter techniques.Experimental results demonstrate that the proposed method significantlyoutperforms the existing single-frame based methods and achieves 13.5%, 11.3%,11.7% BD-Rate saving on average for AI, LDP and RA configurations,respectively, compared with the HEVC reference software.

A Global Appearance and Local Coding Distortion based Fusion Framework for CNN based Filtering in Video Coding