TRACE: Structure-Aware Character Encoding for Robust and Generalizable Document Watermarking

TL;DR

TRACE introduces a structure-aware character encoding framework using diffusion models to embed data into text, offering robustness against noise and broad applicability across languages and fonts.

Contribution

The paper presents a novel diffusion-based encoding method that leverages character structures for robust, generalizable document watermarking, surpassing existing approaches.

Findings

Over 5 dB PSNR improvement compared to state-of-the-art methods.

5% higher extraction accuracy after cross-media transmission.

Effective across multiple languages and fonts.

Abstract

We propose TRACE, a structure-aware framework leveraging diffusion models for localized character encoding to embed data. Unlike existing methods that rely on edge features or pre-defined codebooks, TRACE exploits character structures that provide inherent resistance to noise interference due to their stability and unified representation across diverse characters. Our framework comprises three key components: (1) adaptive diffusion initialization that automatically identifies handle points, target points, and editing regions through specialized algorithms including movement probability estimator (MPE), target point estimation (TPE) and mask drawing model (MDM), (2) guided diffusion encoding for precise movement of selected point, and (3) masked region replacement with a specialized loss function to minimize feature alterations after the diffusion process. Comprehensive experiments demonstrate \name{}'s superior performance over state-of-the-art methods, achieving more than 5 dB improvement in PSNR and 5\% higher extraction accuracy following cross-media transmission. \name{} achieves broad generalizability across multiple languages and fonts, making it particularly suitable for practical document security applications.