# Harnessing Frequency Spectrum Insights for Image Copyright Protection Against Diffusion Models

Source: [Harnessing Frequency Spectrum Insights for Image Copyright Protection Against Diffusion Models](<https://www.sota2.com/research/paper/68fa77721e209131cea73257>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:20:02.162Z

Authors: Zhenguang Liu, Chao Shuai, Shaojing Fan, Ziping Dong, Jinwu Hu, Zhongjie Ba, Kui Ren

Published: 2025-03-14T04:27:50.000Z

## Abstract

Diffusion models have achieved remarkable success in novel view synthesis, but their reliance on large, diverse, and often untraceable Web datasets has raised pressing concerns about image copyright protection. Current methods fall short in reliably identifying unauthorized image use, as they struggle to generalize across varied generation tasks and fail when the training dataset includes images from multiple sources with few identifiable (watermarked or poisoned) samples. In this paper, we present novel evidence that diffusion-generated images faithfully preserve the statistical properties of their training data, particularly reflected in their spectral features. Leveraging this insight, we introduce \\emph{CoprGuard}, a robust frequency domain watermarking framework to safeguard against unauthorized image usage in diffusion model training and fine-tuning. CoprGuard demonstrates remarkable effectiveness against a wide range of models, from naive diffusion models to sophisticated text-to-image models, and is robust even when watermarked images comprise a mere 1\\% of the training dataset. This robust and versatile approach empowers content owners to protect their intellectual property in the era of AI-driven image generation.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Quality Assessment on FFHQ](<https://www.sota2.com/research/sota/image-quality-assessment-on-ffhq.md>): Image Quality Assessment; FFHQ; PSNR: 38.02.

- [Infringement Model Detection on Pokemon](<https://www.sota2.com/research/sota/infringement-model-detection-on-pokemon.md>): Infringement Model Detection; Pokemon; TP: 25.

- [Infringement Model Detection on ImageNet](<https://www.sota2.com/research/sota/infringement-model-detection-on-imagenet.md>): Infringement Model Detection; ImageNet; TP: 10.

- [Image Quality Assessment on ImageNet (PSNR/SSIM)](<https://www.sota2.com/research/sota/image-quality-assessment-on-imagenet-psnr-ssim.md>): Image Quality Assessment; ImageNet; PSNR: 39.91.

- [Image Quality Assessment on LSUN](<https://www.sota2.com/research/sota/image-quality-assessment-on-lsun.md>): Image Quality Assessment; LSUN; PSNR: 34.93.

- [Image Quality Assessment on Pokemon](<https://www.sota2.com/research/sota/image-quality-assessment-on-pokemon.md>): Image Quality Assessment; Pokemon; PSNR: 35.72.

- [Infringement Model Detection on FFHQ](<https://www.sota2.com/research/sota/infringement-model-detection-on-ffhq.md>): Infringement Model Detection; FFHQ; True Positives: 10.

- [Watermark Detection on ImageNet](<https://www.sota2.com/research/sota/watermark-detection-on-imagenet.md>): Watermark Detection; ImageNet; Pu (R=0): 0.

- [Watermark Detection on Pokemon](<https://www.sota2.com/research/sota/watermark-detection-on-pokemon.md>): Watermark Detection; Pokemon; Pu (R=0): 0.

- [Watermark Detection on FFHQ](<https://www.sota2.com/research/sota/watermark-detection-on-ffhq.md>): Watermark Detection; FFHQ; Pu (R=0): 0.

- [Watermark Detection on CUB-200](<https://www.sota2.com/research/sota/watermark-detection-on-cub-200.md>): Watermark Detection; CUB-200; Pu (R=0): not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2503.11071>)

[PDF](<https://arxiv.org/pdf/2503.11071.pdf>)

[Implementation](<https://github.com/sccsok/CoprGuard.git>)