English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

With the spread of tampered images, locating the tampered regions in digitalimages has drawn increasing attention. The existing image tamperinglocalization methods, however, suffer from severe performance degradation whenthe tampered images are subjected to some post-processing, as the tamperingtraces would be distorted by the post-processing operations. The poorrobustness against post-processing has become a bottleneck for the practicalapplications of image tampering localization techniques. In order to addressthis issue, this paper proposes a novel restoration-assisted framework forimage tampering localization (ReLoc). The ReLoc framework mainly consists of animage restoration module and a tampering localization module. The key idea ofReLoc is to use the restoration module to recover a high-quality counterpart ofthe distorted tampered image, such that the distorted tampering traces can bere-enhanced, facilitating the tampering localization module to identify thetampered regions. To achieve this, the restoration module is optimized not onlywith the conventional constraints on image visual quality but also with aforensics-oriented objective function. Furthermore, the restoration module andthe localization module are trained alternately, which can stabilize thetraining process and is beneficial for improving the performance. The proposedframework is evaluated by fighting against JPEG compression, the most commonlyused post-processing. Extensive experimental results show that ReLoc cansignificantly improve the robustness against JPEG compression. The restorationmodule in a well-trained ReLoc model is transferable. Namely, it is stilleffective when being directly deployed with another tampering localizationmodule.

ReLoc: A Restoration-Assisted Framework for Robust Image Tampering Localization