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

Research in presentation attack detection (PAD) for iris recognition haslargely moved beyond evaluation in "closed-set" scenarios, to emphasize abilityto generalize to presentation attack types not present in the training data.This paper offers several contributions to understand and extend thestate-of-the-art in open-set iris PAD. First, it describes the mostauthoritative evaluation to date of iris PAD. We have curated the largestpublicly-available image dataset for this problem, drawing from 26 benchmarkspreviously released by various groups, and adding 150,000 images being releasedwith the journal version of this paper, to create a set of 450,000 imagesrepresenting authentic iris and seven types of presentation attack instrument(PAI). We formulate a leave-one-PAI-out evaluation protocol, and show that eventhe best algorithms in the closed-set evaluations exhibit catastrophic failureson multiple attack types in the open-set scenario. This includes algorithmsperforming well in the most recent LivDet-Iris 2020 competition, which may comefrom the fact that the LivDet-Iris protocol emphasizes sequestered imagesrather than unseen attack types. Second, we evaluate the accuracy of fiveopen-source iris presentation attack algorithms available today, one of whichis newly-proposed in this paper, and build an ensemble method that beats thewinner of the LivDet-Iris 2020 by a substantial margin. This paper demonstratesthat closed-set iris PAD, when all PAIs are known during training, is a solvedproblem, with multiple algorithms showing very high accuracy, while open-setiris PAD, when evaluated correctly, is far from being solved. The newly-createddataset, new open-source algorithms, and evaluation protocol, made publiclyavailable with the journal version of this paper, provide the experimentalartifacts that researchers can use to measure progress on this importantproblem.

State Of The Art In Open-Set Iris Presentation Attack Detection