FlowMark: Mask-Guided Video Watermarking
2026-07-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed FlowMark, a system that hides watermarks in videos by automatically finding the best areas to embed them, without needing manual input. Their method is designed to keep the watermark hidden well even after the video is compressed or edited, such as frames being swapped or removed. FlowMark also avoids flickering effects so the watermark isn’t noticeable while watching. Tests showed it can reliably hide meaningful 128-bit messages with good visual quality, which helps confirm where and when a video came from or if it has been tampered with.
video watermarkingmask predictorrobustnesscompressiongeometric transformationstemporal editsPSNRcontent provenancedeep learningend-to-end training
Authors
Vishal Asnani, Shruti Agarwal, John Collomosse
Abstract
We present FlowMark, a video watermarking framework guided by automatically predicted object masks. In contrast to prior region-based approaches that require user-supplied mask guidance, FlowMark learns to identify optimal regions for watermark embedding through a dedicated Mask Predictor network. Our end-to-end trainable architecture combines region-aware encoding with noise-augmented training to ensure robustness against compression, geometric transformations, and content variation, while preserving high perceptual quality. Our content-adaptive masking keeps watermark signals coherent with natural video dynamics, effectively eliminating perceptual flicker. Beyond compression robustness, FlowMark maintains reliable watermark recovery under video-native temporal edits (e.g., frame swap, insertion, deletion, resampling, and interpolation) and real-world social media distribution pipelines (e.g., YouTube and Facebook re-encoding). Experimental results on both image and video datasets show that FlowMark reliably embeds $128$-bit messages with up to $50.08$ dB PSNR, offering strong performance for content provenance, temporal authenticity verification, and video integrity protection.