On the Coexistence and Ensembling of Watermarks

Aleksandar Petrov, Shruti Agarwal, Philip Torr, Adel Bibi, John Collomosse

Keywords: Watermarking, Ensembling, Machine Learning

TL;DR: We find that, surprisingly, different watermarks can coexist in the same image which enables us to build ensembles of watermarks that can open up new accuracy-robustness-capcity-quality trade-offs without further training.

Abstract: Watermarking, the practice of embedding imperceptible information into media such as images, videos, audio, and text, is essential for intellectual property protection, content provenance and attribution. The growing complexity of digital ecosystems necessitates watermarks for different uses to be embedded in the same media. However, to detect and decode all watermarks, they need to coexist well with one another. We perform the first study of coexistence of deep image watermarking methods and, contrary to intuition, we find that various open-source watermarks can coexist with only minor impacts on image quality and decoding robustness. The coexistence of watermarks also opens the avenue for ensembling watermarking methods. We show how ensembling can increase the overall message capacity and enable new trade-offs between capacity, accuracy, robustness and image quality, without needing to retrain the base models.

Primary Area: Social and economic aspects of machine learning (e.g., fairness, interpretability, human-AI interaction, privacy, safety, strategic behavior)

Submission Number: 13914