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

Despite the tremendous success, deep neural networks are exposed to seriousIP infringement risks. Given a target deep model, if the attacker knows itsfull information, it can be easily stolen by fine-tuning. Even if only itsoutput is accessible, a surrogate model can be trained through student-teacherlearning by generating many input-output training pairs. Therefore, deep modelIP protection is important and necessary. However, it is still seriouslyunder-researched. In this work, we propose a new model watermarking frameworkfor protecting deep networks trained for low-level computer vision or imageprocessing tasks. Specifically, a special task-agnostic barrier is added afterthe target model, which embeds a unified and invisible watermark into itsoutputs. When the attacker trains one surrogate model by using the input-outputpairs of the barrier target model, the hidden watermark will be learned andextracted afterwards. To enable watermarks from binary bits to high-resolutionimages, a deep invisible watermarking mechanism is designed. By jointlytraining the target model and watermark embedding, the extra barrier can evenbe absorbed into the target model. Through extensive experiments, wedemonstrate the robustness of the proposed framework, which can resist attackswith different network structures and objective functions.

Deep Model Intellectual Property Protection via Deep Watermarking