OSI: One-step Inversion Excels in Extracting Diffusion Watermarks

2026-02-10

AI Summary

This work addresses the high computational cost and inefficiency of traditional diffusion-based watermark extraction, which relies on multi-step inversion. The authors propose the first approach that reformulates watermark extraction as a symbol classification task. By fine-tuning the diffusion backbone model on synthetic noise–image pairs, their method enables single-step, highly efficient extraction without requiring precise reconstruction of the initial noise. This paradigm shift yields substantial performance gains: the approach achieves approximately 20× faster extraction and doubles watermark capacity compared to existing methods, while maintaining high accuracy and robustness across diverse schedulers, backbone architectures, and encryption schemes.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Information Extraction

Application Category

• Web Mining and Content Analysis: Content-based information diffusion
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Watermarking is an important mechanism for provenance and copyright protection of diffusion-generated images. Training-free methods, exemplified by Gaussian Shading, embed watermarks into the initial noise of diffusion models with negligible impact on the quality of generated images. However, extracting this type of watermark typically requires multi-step diffusion inversion to obtain precise initial noise, which is computationally expensive and time-consuming. To address this issue, we propose One-step Inversion (OSI), a significantly faster and more accurate method for extracting Gaussian Shading style watermarks. OSI reformulates watermark extraction as a learnable sign classification problem, which eliminates the need for precise regression of the initial noise. Then, we initialize the OSI model from the diffusion backbone and finetune it on synthesized noise-image pairs with a sign classification objective. In this manner, the OSI model is able to accomplish the watermark extraction efficiently in only one step. Our OSI substantially outperforms the multi-step diffusion inversion method: it is 20x faster, achieves higher extraction accuracy, and doubles the watermark payload capacity. Extensive experiments across diverse schedulers, diffusion backbones, and cryptographic schemes consistently show improvements, demonstrating the generality of our OSI framework.

Problem

• watermark extraction
• diffusion models
• diffusion inversion
• computational efficiency
• initial noise

Innovation

• One-step Inversion
• Diffusion Watermarking
• Sign Classification
• Training-free Watermark
• Fast Watermark Extraction