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

Recent deep learning based video synthesis approaches, in particular withapplications that can forge identities such as "DeepFake", have raised greatsecurity concerns. Therefore, corresponding deep forensic methods are proposedto tackle this problem. However, existing methods are either based onunexplainable deep networks which greatly degrades the principalinterpretability factor to media forensic, or rely on fragile image statisticssuch as noise pattern, which in real-world scenarios can be easily deterioratedby data compression. In this paper, we propose an fully-interpretable videoforensic method that is designed specifically to expose deep-faked videos. Toenhance generalizability on videos with various content, we model the temporalmotion of multiple specific spatial locations in the videos to extract a robustand reliable representation, called Co-Motion Pattern. Such kind of conjointpattern is mined across local motion features which is independent of the videocontents so that the instance-wise variation can also be largely alleviated.More importantly, our proposed co-motion pattern possesses both superiorinterpretability and sufficient robustness against data compression fordeep-faked videos. We conduct extensive experiments to empirically demonstratethe superiority and effectiveness of our approach under both classification andanomaly detection evaluation settings against the state-of-the-art deepforensic methods.

Exposing Deep-faked Videos by Anomalous Co-motion Pattern Detection