ADD for Multi-Bit Image Watermarking

TL;DR

This paper introduces ADD, a novel multi-bit image watermarking method that achieves high accuracy and robustness with significant computational efficiency, addressing limitations of existing approaches.

Contribution

ADD provides a new linear watermarking technique with theoretical justification, improving capacity, robustness, and speed over prior methods.

Findings

Achieves 100% decoding accuracy for 48-bit watermarking on MS-COCO.

Maintains performance with at most 2% drop under various image distortions.

Embeds and decodes images 2 to 7.4 times faster than existing methods.

Abstract

As generative models enable rapid creation of high-fidelity images, societal concerns about misinformation and authenticity have intensified. A promising remedy is multi-bit image watermarking, which embeds a multi-bit message into an image so that a verifier can later detect whether the image is generated by someone and further identify the source by decoding the embedded message. Existing approaches often fall short in capacity, resilience to common image distortions, and theoretical justification. To address these limitations, we propose ADD (Add, Dot, Decode), a multi-bit image watermarking method with two stages: learning a watermark to be linearly combined with the multi-bit message and added to the image, and decoding through inner products between the watermarked image and the learned watermark. On the standard MS-COCO benchmark, we demonstrate that for the challenging task of 48-bit watermarking, ADD achieves 100\% decoding accuracy, with performance dropping by at most 2\% under a wide range of image distortions, substantially smaller than the 14\% average drop of state-of-the-art methods. In addition, ADD achieves substantial computational gains, with 2-fold faster embedding and 7.4-fold faster decoding than the fastest existing method. We further provide a theoretical analysis explaining why the learned watermark and the corresponding decoding rule are effective.