DeepSignature: Digitally Signed, Content-Encoding Watermarks for Robust and Transparent Image Authentication

2026-04-24

AI Summary

This study addresses the credibility crisis precipitated by AI-generated imagery by proposing a novel approach that integrates cryptographic digital signatures with deep learning-based watermarking. The method imperceptibly embeds content-bound watermarks into standard image formats, enabling verifiable source attribution and content integrity authentication. Its key innovation lies in the first-time fusion of provably secure cryptographic signatures with neural network watermarking, facilitating lightweight client-side verification and incorporating a latent-space tampering localization mechanism. Experimental results demonstrate that the technique achieves near-perfect detection rates—approaching 100%—against prevalent forgery attacks while maintaining high perceptual invisibility, thereby striking an effective balance among robustness, security, and compatibility with existing image standards.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

AI-powered generative models have significantly expanded the possibilities for editing, manipulating, and creating high-quality images. Particularly, images that falsely appear to originate from trusted sources pose a serious threat, undermining public trust in image authenticity. We propose DeepSignature, a novel approach that integrates the guarantees of digital signatures with the capabilities of deep neural networks. Neural networks are used both to generate content-encoding watermarks and to embed them imperceptibly into images while ensuring robust extraction. These watermarks are cryptographically verifiable, enabling source attribution and image integrity validation. DeepSignature is compatible with existing image formats and requires no special handling of signed images. It supports client-side verification, requiring only the signer's public key. Additionally, we introduce a novel latent-space verification approach to detect and localize tampering attempts. We evaluate DeepSignature in terms of imperceptibility, robustness to benign transformations, forgery detection, and its resilience against various attack scenarios. Our results highlight the inherent trade-offs between imperceptibility, robustness, and integrity verification. We demonstrate that DeepSignature reliably identifies significant forgery attempts -- achieving near 100\% in our experiments. Finally, we emphasize DeepSignature's modularity and tunable parameters, allowing adaptation to application-specific requirements. Code and model weights will be published.

Problem

• image authentication
• digital watermarking
• generative AI
• content integrity
• source attribution

Innovation

• digital watermarking
• image authentication
• deep neural networks
• cryptographic verification
• tamper localization