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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Face presentation attack detection (PAD) is an essential measure to protectface recognition systems from being spoofed by malicious users and hasattracted great attention from both academia and industry. Although most of theexisting methods can achieve desired performance to some extent, thegeneralization issue of face presentation attack detection under cross-domainsettings (e.g., the setting of unseen attacks and varying illumination) remainsto be solved. In this paper, we propose a novel framework based on asymmetricmodality translation for face presentation attack detection in bi-modalityscenarios. Under the framework, we establish connections between two modalityimages of genuine faces. Specifically, a novel modality fusion scheme ispresented that the image of one modality is translated to the other one throughan asymmetric modality translator, then fused with its corresponding pairedimage. The fusion result is fed as the input to a discriminator for inference.The training of the translator is supervised by an asymmetric modalitytranslation loss. Besides, an illumination normalization module based onPattern of Local Gravitational Force (PLGF) representation is used to reducethe impact of illumination variation. We conduct extensive experiments on threepublic datasets, which validate that our method is effective in detectingvarious types of attacks and achieves state-of-the-art performance underdifferent evaluation protocols.

Asymmetric Modality Translation For Face Presentation Attack Detection