Towards Photo-Realistic Visible Watermark Removal with Conditional Generative Adversarial Networks

Abstract

Visible watermark plays an important role in image copyright protection and the robustness of a visible watermark to an attack is shown to be essential. To evaluate and improve the effectiveness of watermark, watermark removal attracts increasing attention and becomes a hot research top. Current methods cast the watermark removal as an image-to-image translation problem where the encode-decode architectures with pixel-wise loss are adopted to transfer the transparent watermarked pixels into unmarked pixels. However, when a number of realistic images are presented, the watermarks are more likely to be unknown and diverse (i.e., the watermarks might be opaque or semi-transparent; the category and pattern of watermarks are unknown). When applying existing methods to the real-world scenarios, they mostly can not satisfactorily reconstruct the hidden information obscured under the complex and various watermarks (i.e., the residual watermark traces remain and the reconstructed images lack reality). To address this difficulty, in this paper, we present a new watermark processing framework using the conditional generative adversarial networks (cGANs) for visible watermark removal in the real-world application. The proposed method enables the watermark removal solution to be more closed to the photo-realistic reconstruction using a patch-based discriminator conditioned on the watermarked images, which is adversarially trained to differentiate the difference between the recovered images and original watermark-free images. Extensive experimental results on a large-scale visible watermark dataset demonstrate the effectiveness of the proposed method and clearly indicate that our proposed approach can produce more photo-realistic and convincing results compared with the state-of-the-art methods.

Visible Watermark Removal via Self-calibrated Localization and Background Refinement

“…One observation is that results on LVW dataset are much better than those on CLWD dataset, because LVW dataset only contains gray-scale watermarks and is much easier than CLWD dataset. Another observation is that the image content removal methods [7,32] and watermark removal methods [6,19,28] based multi-task learning outperform image-to-image translation method [24] by a large margin, which verifies the effectiveness and necessity of predicting watermark mask. Moreover, baselines SplitNet [6], BVMR [19] and WDNet [28] specifically designed for watermark removal perform more favorably on two datasets than image content removal methods [7,32].…”

“…The development of deep learning techniques have greatly advanced the watermark removal task. Some methods [2,24] formulated the watermark removal as an image-to-image translation task. Other methods [6,19,28] performed watermark localization and removal tasks at the same time.…”

“…However, watermarked images are likely to be converted back to watermarkfree images by virtue of modern watermark removal techniques. To evaluate and strengthen the robustness of visible watermarks in an adversarial way, watermark removal task has raised the research interest in recent years [2,4,6,9,19,24].…”

“…Unfortunately, the assumption in [9,15] limits their application to real-world scenarios. Recently, researchers [2,4,6,19,24,28] attempt to solve the blind watermark removal problem in an end-to-end manner with deep learning approaches. Some works [2,24] formulated the watermark removal problem as an image-to-image translation task without localizing the watermark.…”

“…Recently, researchers [2,4,6,19,24,28] attempt to solve the blind watermark removal problem in an end-to-end manner with deep learning approaches. Some works [2,24] formulated the watermark removal problem as an image-to-image translation task without localizing the watermark. On the contrary, other works realized that watermark should be localized and removed sequentially [4] or simultaneously [6,19,28].…”