SpecGuard: Spectral Projection-based Advanced Invisible Watermarking

2025-10-08

AI Summary

Existing image watermarking methods exhibit insufficient robustness against distortions, resampling, and adversarial perturbations. This paper proposes a robust, imperceptible watermarking framework based on spectral projection. First, the host image undergoes wavelet decomposition to isolate high-frequency subbands; these are then mapped into the frequency domain via fast Fourier transform (FFT). Watermark embedding is performed within the hidden layers of a convolutional neural network. Crucially, the method enforces Parseval’s theorem to constrain both encoder and decoder, enabling energy-preserving, reversible projections in the frequency domain—thereby jointly optimizing imperceptibility and robustness. Experiments demonstrate state-of-the-art performance across multiple metrics: higher PSNR and SSIM values, greater watermark capacity, and superior extraction accuracy under diverse attacks—including JPEG compression, cropping, additive noise, and adversarial perturbations. The source code is publicly available.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Measurement, analysis, and circumvention of Web censorship
• Security and Privacy: Large-scale security measurements

Abstract

Watermarking embeds imperceptible patterns into images for authenticity verification. However, existing methods often lack robustness against various transformations primarily including distortions, image regeneration, and adversarial perturbation, creating real-world challenges. In this work, we introduce SpecGuard, a novel watermarking approach for robust and invisible image watermarking. Unlike prior approaches, we embed the message inside hidden convolution layers by converting from the spatial domain to the frequency domain using spectral projection of a higher frequency band that is decomposed by wavelet projection. Spectral projection employs Fast Fourier Transform approximation to transform spatial data into the frequency domain efficiently. In the encoding phase, a strength factor enhances resilience against diverse attacks, including adversarial, geometric, and regeneration-based distortions, ensuring the preservation of copyrighted information. Meanwhile, the decoder leverages Parseval's theorem to effectively learn and extract the watermark pattern, enabling accurate retrieval under challenging transformations. We evaluate the proposed SpecGuard based on the embedded watermark's invisibility, capacity, and robustness. Comprehensive experiments demonstrate the proposed SpecGuard outperforms the state-of-the-art models. To ensure reproducibility, the full code is released on href{https://github.com/inzamamulDU/SpecGuard_ICCV_2025}{ extcolor{blue}{ extbf{GitHub}}}.

Problem

• Develops robust invisible watermarking against image distortions
• Embeds messages in frequency domain using spectral projection
• Enhances resilience to adversarial and geometric attacks

Innovation

• Embeds watermarks in hidden convolution layers
• Uses spectral projection with Fast Fourier Transform
• Employs Parseval's theorem for watermark extraction