Search papers, labs, and topics across Lattice.
This paper introduces Watermark Removal Detection (WRD) as a critical, yet overlooked, evaluation metric for watermark removal methods, highlighting that current methods leave detectable statistical artifacts even when achieving high attack success and perceptual quality. They demonstrate that a classifier trained on these artifacts achieves state-of-the-art detection rates, exposing the forensic vulnerability of existing removal techniques. The authors benchmark leading watermarking schemes and removal pipelines, revealing that no current method effectively balances attack success, perceptual quality, and forensic stealthiness.
Watermark removal methods may fool the eye, but they leave behind statistical fingerprints that are easily detectable by a forensic classifier.
Current watermark removal methods are evaluated on two axes: attack success rate and perceptual quality. We show this is insufficient. While state-of-the-art attacks successfully degrade the watermark signal without visible distortion, they leave distinct statistical artifacts that betray the removal attempt. We name this overlooked axis Watermark Removal Detection (WRD) and demonstrate that a modern classifier trained on these artifacts achieves state-of-the-art detection rates at $10^{-3}$ FPR across every removal method tested. No existing attack accounts for this forensic leakage. We benchmark leading watermarking schemes against standard removal pipelines under the extended evaluation triple of attack success, perceptual quality, and forensic detectability, and find that no current method balances all three. Our results establish forensic stealthiness as a necessary requirement for watermark removal.
