ShapeMark: Robust and Diversity-Preserving Watermarking for Diffusion Models

TL;DR

ShapeMark introduces a novel watermarking technique for diffusion models that encodes watermarks into structured noise patterns, achieving high robustness and preserving diversity in generated images, thus enhancing intellectual property protection.

Contribution

The paper proposes a new Noise-as-Watermark method that encodes watermarks into structured noise patterns and employs randomization to improve robustness and diversity in diffusion model outputs.

Findings

Achieves state-of-the-art robustness in watermarking diffusion models.

Maintains high image quality and diversity despite lossy scenarios.

Outperforms existing Noise-as-Watermark approaches in experiments.

Abstract

Diffusion models have made substantial advances in recent years, enabling high-quality image synthesis; however, the widespread dissemination and reuse of their outputs have introduced new challenges in intellectual property protection and content provenance. Image watermarking offers a solution to these challenges, and recent work has increasingly explored Noise-as-Watermark (NaW) approaches that integrate watermarking directly into the diffusion process. However, existing NaW methods fail to balance robustness and diversity. We attribute this weakness to value encoding, which encodes watermark bits into individual sampled values. It is extremely fragile in practical application scenarios. To address this, we encode watermark bits into the structured noise pattern, so that the watermark is preserved even when individual values are perturbed. To further ensure generation diversity, we introduce a dedicated randomization design that reshuffles the positions of noise elements without changing their values, preventing the watermark from inducing fixed noise patterns or spatial locations. Extensive experiments demonstrate that our method achieves state-of-the-art robustness while maintaining high generation quality across a wide range of lossy scenarios.