Toward Real-world Text Image Forgery Localization: Structured and Interpretable Data Synthesis

TL;DR

This paper introduces Fourier Series-based Tampering Synthesis (FSTS), a structured data generation framework that improves the realism and diversity of training data for text image forgery localization, enhancing model generalization to real-world cases.

Contribution

FSTS is a novel, interpretable framework that models tampering behaviors hierarchically using Fourier series principles to synthesize realistic training data for forgery localization.

Findings

Models trained with FSTS data show improved generalization on real-world datasets.

FSTS effectively captures complex tampering behaviors through hierarchical modeling.

Synthetic data generated by FSTS enhances the robustness of forgery detection models.

Abstract

Existing Text Image Forgery Localization (T-IFL) methods often suffer from poor generalization due to the limited scale of real-world datasets and the distribution gap caused by synthetic data that fails to capture the complexity of real-world tampering. To tackle this issue, we propose Fourier Series-based Tampering Synthesis (FSTS), a structured and interpretable framework for synthesizing tampered text images. FSTS first collects 16,750 real-world tampering instances from five representative tampering types, using a structured pipeline that records human-performed editing traces via multi-format logs (e.g., video, PSD, and editing logs). By analyzing these collected parameters and identifying recurring behavioral patterns at both individual and population levels, we formulate a hierarchical modeling framework. Specifically, each individual tampering parameter is represented as a compact combination of basis operation-parameter configurations, while the population-level distribution is constructed by aggregating these behaviors. Since this formulation draws inspiration from the Fourier series, it enables an interpretable approximation using basis functions and their learned weights. By sampling from this modeled distribution, FSTS synthesizes diverse and realistic training data that better reflect real-world forgery traces. Extensive experiments across four evaluation protocols demonstrate that models trained with FSTS data achieve significantly improved generalization on real-world datasets. Dataset is available at \href{https://github.com/ZeqinYu/FSTS}{Project Page}.