RAW: Robust Avatar Watermarking -- Benchmarking and Baseline
For digital avatar watermarking, this work provides a benchmark and a baseline method that addresses real-world post-processing attacks.
The paper introduces a benchmark for robust avatar watermarking and proposes WALT, which embeds watermarks in UV texture space, achieving 92.4% robustness to zoom attacks and 95.6% on background removal.
Digital avatar watermarking presents unique challenges: avatars are routinely post-processed with background replacement, reframing, and format conversion before deployment. We introduce \textbf{RAW} (Robust Avatar Watermarking), a benchmark comprising 50 synthetic avatar videos from 5 commercial providers and 6 attacks simulating real-world avatar workflows. Evaluating 7 existing methods reveals that avatar-specific attacks such as background removal significantly degrade watermark recovery. We propose \textbf{WALT} (Watermarking Avatars with Learned Textures), which embeds watermarks in UV texture space via 3D face reconstruction. WALT achieves the highest robustness to zoom attacks (92.4\%) while maintaining strong performance on background removal (95.6\%). We release our benchmark to facilitate research into avatar-specific watermarking.