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 this paper, we study a new problem arising from the emerging MPEGstandardization effort Video Coding for Machine (VCM), which aims to bridge thegap between visual feature compression and classical video coding. VCM iscommitted to address the requirement of compact signal representation for bothmachine and human vision in a more or less scalable way. To this end, we makeendeavors in leveraging the strength of predictive and generative models tosupport advanced compression techniques for both machine and human vision taskssimultaneously, in which visual features serve as a bridge to connectsignal-level and task-level compact representations in a scalable manner.Specifically, we employ a conditional deep generation network to reconstructvideo frames with the guidance of learned motion pattern. By learning toextract sparse motion pattern via a predictive model, the network elegantlyleverages the feature representation to generate the appearance of to-be-codedframes via a generative model, relying on the appearance of the coded keyframes. Meanwhile, the sparse motion pattern is compact and highly effectivefor high-level vision tasks, e.g. action recognition. Experimental resultsdemonstrate that our method yields much better reconstruction quality comparedwith the traditional video codecs (0.0063 gain in SSIM), as well asstate-of-the-art action recognition performance over highly compressed videos(9.4% gain in recognition accuracy), which showcases a promising paradigm ofcoding signal for both human and machine vision.

An Emerging Coding Paradigm VCM: A Scalable Coding Approach Beyond Feature and Signal