Receipt Replay OOD: A Small Benchmark for Screen Replay Detection Under Domain Shift
For researchers in presentation attack detection, this work provides a small benchmark to evaluate OOD robustness, but it is incremental as it only applies existing methods to a new dataset.
The paper introduces Receipt Replay OOD, a small benchmark for evaluating out-of-domain robustness of screen replay detection models, and demonstrates that domain shift significantly impacts generalization performance.
Public datasets such as DLC-2021, SynID, and KID34K have significantly contributed to research on presentation attack detection for identity documents, including screen replay attacks. However, evaluation of out-of-domain (OOD) robustness remains insufficiently explored, especially under realistic domain shifts. In this work, we introduce Receipt Replay OOD, a small out-of-domain benchmark for screen replay detection. Receipts share several characteristics with identity documents, including planar geometry, curved corners, wear-and-tear artifacts, and text or logo patterns, while avoiding personally identifiable information constraints commonly associated with identity documents. We evaluate document replay detection models under cross-domain conditions and demonstrate the impact of domain shift on generalization performance. The dataset is publicly available.