2025/10/08 by Pragati Shuddhodhan Meshram, Meshram, Pragati Shuddhodhan, Chandrasekaran, Varun
Computer Science · #Advanced Steganography and Watermarking Techniques #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis
paper · pdf · doi:10.48550/arxiv.2510.07538
openalex publication_date 2025/10/08 · openalex created_date 2025/10/11 · openalex updated_date 2026/07/28
Effective removal of semantic watermarks requires balancing three competing objectives: high removal success, low perceptual distortion, and low computational cost. However, existing single-image attacks typically optimize only for the first two, achieving strong watermark suppression but relying on expensive, multi-step optimization that limits practical deployment. In this work, we show that this trade-off is fundamental: no current approach achieves all three properties simultaneously. We introduce DAWN, a lightweight, training-free attack that explicitly targets the low-cost regime while maintaining competitive removal performance. DAWN works by projecting a watermarked image onto natural-image priors in complementary frequency and semantic spaces, suppressing watermark signals that deviate from natural statistics, and then applying a decoupled perceptual-alignment step to restore visual consistency with minimal artifact. Across diverse pixel-, frequency-, and latent-space watermarking schemes, DAWN consistently reduces detectability while preserving structural and semantic fidelity, demonstrating that efficient, low-resource watermark removal is feasible with only modest perceptual degradation. Our code is available at https://github.com/Pragati-Meshram/DAWN.