Go Wide or Go Deep: Levering Watermarking Performance with Computational Cost for Specific Images

TL;DR

This paper introduces ISMark, a novel image-specific watermarking framework that customizes embedding mechanisms for individual images, significantly enhancing robustness and visual quality by leveraging computation cost and optimization techniques.

Contribution

The paper proposes a new auto-decoder-like watermarking framework that breaks the robustness-invisibility trade-off by optimizing cover images for each specific case.

Findings

ISMark outperforms state-of-the-art methods in robustness and visual quality.

It achieves a 4.64% improvement in bit error rate.

It increases PSNR by 2.20dB, indicating better visual quality.

Abstract

Digital watermarking has been widely studied for the protection of intellectual property. Traditional watermarking schemes often design in a "wider" rule, which applies one general embedding mechanism to all images. But this will limit the scheme into a robustness-invisibility trade-off, where the improvements of robustness can only be achieved by the increase of embedding intensity thus causing the visual quality decay. However, a new scenario comes out at this stage that many businesses wish to give high level protection to specific valuable images, which requires high robustness and high visual quality at the same time. Such scenario makes the watermarking schemes should be designed in a "deeper" way which makes the embedding mechanism customized to specific images. To achieve so, we break the robustness-invisibility trade-off by introducing computation cost in, and propose a novel auto-decoder-like image-specified watermarking framework (ISMark). Based on ISMark, the strong robustness and high visual quality for specific images can be both achieved. In detail, we apply an optimization procedure (OPT) to replace the traditional embedding mechanism. Unlike existing schemes that embed watermarks using a learned encoder, OPT regards the cover image as the optimizable parameters to minimize the extraction error of the decoder, thus the features of each specified image can be effectively exploited to achieve superior performance. Extensive experiments indicate that ISMark outperforms the state-of-the-art methods by a large margin, which improves the average bit error rate by 4.64% (from 4.86% to 0.22%) and PSNR by 2.20dB (from 32.50dB to 34.70dB).