Watermarking in Diffusion Model: Gaussian Shading with Exact Diffusion Inversion via Coupled Transformations (EDICT)

TL;DR

This paper presents a novel method combining EDICT with Gaussian Shading to improve the accuracy and robustness of watermark recovery in diffusion-based image watermarking techniques.

Contribution

It introduces the first integration of EDICT's exact inversion capabilities with Gaussian Shading, enhancing watermark reconstruction fidelity in diffusion models.

Findings

Improved watermark recovery fidelity demonstrated on standard datasets.

Statistically significant enhancement over traditional Gaussian Shading methods.

First exploration of EDICT's application in diffusion-based watermarking.

Abstract

This paper introduces a novel approach to enhance the performance of Gaussian Shading, a prevalent watermarking technique, by integrating the Exact Diffusion Inversion via Coupled Transformations (EDICT) framework. While Gaussian Shading traditionally embeds watermarks in a noise latent space, followed by iterative denoising for image generation and noise addition for watermark recovery, its inversion process is not exact, leading to potential watermark distortion. We propose to leverage EDICT's ability to derive exact inverse mappings to refine this process. Our method involves duplicating the watermark-infused noisy latent and employing a reciprocal, alternating denoising and noising scheme between the two latents, facilitated by EDICT. This allows for a more precise reconstruction of both the image and the embedded watermark. Empirical evaluation on standard datasets demonstrates that our integrated approach yields a slight, yet statistically significant improvement in watermark recovery fidelity. These results highlight the potential of EDICT to enhance existing diffusion-based watermarking techniques by providing a more accurate and robust inversion mechanism. To the best of our knowledge, this is the first work to explore the synergy between EDICT and Gaussian Shading for digital watermarking, opening new avenues for research in robust and high-fidelity watermark embedding and extraction.