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

Offline Handwritten Signature verification presents a challenging PatternRecognition problem, where only knowledge of the positive class is availablefor training. While classifiers have access to a few genuine signatures fortraining, during generalization they also need to discriminate forgeries. Thisis particularly challenging for skilled forgeries, where a forger practicesimitating the user's signature, and often is able to create forgeries visuallyclose to the original signatures. Most work in the literature address thisissue by training for a surrogate objective: discriminating genuine signaturesof a user and random forgeries (signatures from other users). In this work, wepropose a solution for this problem based on meta-learning, where there are twolevels of learning: a task-level (where a task is to learn a classifier for agiven user) and a meta-level (learning across tasks). In particular, themeta-learner guides the adaptation (learning) of a classifier for each user,which is a lightweight operation that only requires genuine signatures. Themeta-learning procedure learns what is common for the classification acrossdifferent users. In a scenario where skilled forgeries from a subset of usersare available, the meta-learner can guide classifiers to be discriminative ofskilled forgeries even if the classifiers themselves do not use skilledforgeries for learning. Experiments conducted on the GPDS-960 dataset showimproved performance compared to Writer-Independent systems, and achieveresults comparable to state-of-the-art Writer-Dependent systems in the regimeof few samples per user (5 reference signatures).

Meta-learning for fast classifier adaptation to new users of Signature Verification systems