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 Signature Verification (OSV) is a fundamental biometric task acrossvarious forensic, commercial and legal applications. The underlying task athand is to carefully model fine-grained features of the signatures todistinguish between genuine and forged ones, which differ only in minutedeformities. This makes OSV more challenging compared to other verificationproblems. In this work, we propose a two-stage deep learning framework thatleverages self-supervised representation learning as well as metric learningfor writer-independent OSV. First, we train an image reconstruction networkusing an encoder-decoder architecture that is augmented by a 2D spatialattention mechanism using signature image patches. Next, the trained encoderbackbone is fine-tuned with a projector head using a supervised metric learningframework, whose objective is to optimize a novel dual triplet loss by samplingnegative samples from both within the same writer class as well as from otherwriters. The intuition behind this is to ensure that a signature sample liescloser to its positive counterpart compared to negative samples from bothintra-writer and cross-writer sets. This results in robust discriminativelearning of the embedding space. To the best of our knowledge, this is thefirst work of using self-supervised learning frameworks for OSV. The proposedtwo-stage framework has been evaluated on two publicly available offlinesignature datasets and compared with various state-of-the-art methods. It isnoted that the proposed method provided promising results outperforming severalexisting pieces of work. The code is publicly available at:https://github.com/soumitri2001/SURDS-SSL-OSV

SURDS: Self-Supervised Attention-guided Reconstruction and Dual Triplet Loss for Writer Independent Offline Signature Verification