Signature Forgery Detection: Improving Cross-Dataset Generalization
This work addresses a critical issue for banking and identity authentication by improving model robustness across different signature datasets, though it appears incremental as it compares existing methods without establishing clear superiority.
This study tackled the problem of poor cross-dataset generalization in offline signature verification by investigating feature learning strategies, finding that a raw-image model performed better across benchmarks while a shell-based approach showed potential for future improvements.
Automated signature verification is a critical biometric technique used in banking, identity authentication, and legal documentation. Despite the notable progress achieved by deep learning methods, most approaches in offline signature verification still struggle to generalize across datasets, as variations in handwriting styles and acquisition protocols often degrade performance. This study investigates feature learning strategies for signature forgery detection, focusing on improving cross-dataset generalization -- that is, model robustness when trained on one dataset and tested on another. Using three public benchmarks -- CEDAR, ICDAR, and GPDS Synthetic -- two experimental pipelines were developed: one based on raw signature images and another employing a preprocessing method referred to as shell preprocessing. Several behavioral patterns were identified and analyzed; however, no definitive superiority between the two approaches was established. The results show that the raw-image model achieved higher performance across benchmarks, while the shell-based model demonstrated promising potential for future refinement toward robust, cross-domain signature verification.