Domain Generalized Recaptured Screen Image Identification Using SWIN Transformer

TL;DR

This paper introduces DAST-DG, a novel SWIN transformer-based framework with data augmentation for domain-generalized recaptured screen image identification, effectively handling scale variations and domain shifts.

Contribution

It proposes a cascaded data augmentation and SWIN transformer approach that improves domain generalization in recaptured image identification, outperforming existing methods.

Findings

Achieves approximately 82% accuracy on high-variance datasets

Attains 95% precision in recaptured image detection

Outperforms state-of-the-art methods across multiple databases

Abstract

An increasing number of classification approaches have been developed to address the issue of image rebroadcast and recapturing, a standard attack strategy in insurance frauds, face spoofing, and video piracy. However, most of them neglected scale variations and domain generalization scenarios, performing poorly in instances involving domain shifts, typically made worse by inter-domain and cross-domain scale variances. To overcome these issues, we propose a cascaded data augmentation and SWIN transformer domain generalization framework (DAST-DG) in the current research work Initially, we examine the disparity in dataset representation. A feature generator is trained to make authentic images from various domains indistinguishable. This process is then applied to recaptured images, creating a dual adversarial learning setup. Extensive experiments demonstrate that our approach is practical and surpasses state-of-the-art methods across different databases. Our model achieves an accuracy of approximately 82\% with a precision of 95\% on high-variance datasets.