A Deep Multiscale Framework for Video Watermarking

원문 초록

Video watermarking embeds a message into a cover video in an imperceptible manner, which can be retrieved even if the video undergoes certain modifications or distor￾tions. Traditional watermarking methods are often manu￾ally designed for particular types of distortions and thus cannot simultaneously handle a broad spectrum of distor￾tions. To this end, we propose a robust deep learning-based solution for video watermarking that is end-to-end train￾able. Our model consists of a novel multiscale design where the watermarks are distributed across multiple spatial￾temporal scales. It gains robustness against various dis￾tortions through a differentiable distortion layer, whereas non-differentiable distortions, such as popular video com￾pression standards, are modeled by a differentiable proxy. Extensive evaluations on a wide variety of distortions show that our method outperforms traditional video watermark￾ing methods as well as deep image watermarking models by a large margin. We further demonstrate the practicality of our method on a realistic video-editing application.