Exploring Vision-Language Models for Online Signature Verification: A Zero-Shot Capability Study

TL;DR

This study evaluates the zero-shot performance of advanced Vision-Language Models on online signature verification, revealing strengths in random forgery detection and challenges with skilled forgeries and reasoning artifacts.

Contribution

It introduces a novel protocol for biometric scoring using VLMs and provides the first exploration of their zero-shot capabilities in signature verification tasks.

Findings

GPT-5.2 achieves 0.32% EER in mobile tasks for random forgeries.

VLMs outperform supervised systems in random forgery scenarios.

Performance drops significantly in skilled forgery scenarios due to reasoning issues.

Abstract

Recent advancements in Vision-Language Models (VLMs) have demonstrated strong capabilities in general visual reasoning, yet their applicability to rigorous biometric tasks remains unexplored. This work presents an exploratory study evaluating the zero-shot performance of state-of-the-art VLMs (GPT-5.2 and Gemini 2.5 Pro) on the Signature Verification Challenge (SVC) benchmark. To enable visual processing, raw kinematic time-series are converted into static images, encoding pressure information into stroke opacity whenever available in the source data. Furthermore, we introduce a scoring protocol that extracts latent token probabilities to compute robust biometric scores. Experimental results reveal a significant performance dichotomy dependent on signal quality and forgery type. In random forgery scenarios, the zero-shot VLM achieves exceptional discrimination, with GPT-5.2 reaching an Equal Error Rate of 0.32% in mobile tasks, outperforming supervised state-of-the-art systems. Conversely, in skilled forgery scenarios, where the task is more challenging because both signatures are almost identical, the results are significantly worse, and a critical "Rationalization Trap" emerges: chain-of-thought (CoT) reasoning degrades performance as the model produces kinematic hallucinations to justify forgery artifacts as natural variability.