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

Rounding confidence score is considered trivial but a simple and effectivecountermeasure to stop gradient descent based image reconstruction attacks.However, its capability in the face of more sophisticated reconstructionattacks is an uninvestigated research area. In this paper, we prove that, theface reconstruction attacks based on composite faces can reveal theinefficiency of rounding policy as countermeasure. We assume that, the attackertakes advantage of face composite parts which helps the attacker to get accessto the most important features of the face or decompose it to the independentsegments. Afterwards, decomposed segments are exploited as search parameters tocreate a search path to reconstruct optimal face. Face composition parts enablethe attacker to violate the privacy of face recognition models even with ablind search. However, we assume that, the attacker may take advantage ofrandom search to reconstruct the target face faster. The algorithm is startedwith random composition of face parts as initial face and confidence score isconsidered as fitness value. Our experiments show that, since the roundingpolicy as countermeasure can't stop the random search process, current facerecognition systems are extremely vulnerable against such sophisticatedattacks. To address this problem, we successfully test Face Detection ScoreFiltering (FDSF) as a countermeasure to protect the privacy of training dataagainst proposed attack.

Vulnerability of Face Recognition Systems Against Composite Face Reconstruction Attack