Adversarial photo frame: Concealing sensitive scene information of social images in a user-acceptable manner

Research output: Contribution to journalConference articlepeer-review

Abstract

Personal privacy protection has become more and more crucial in the era of big multimedia data and artificial intelligence. This paper presents our submission to pixel privacy task, where we propose to fool the deep visual classification model that is for recognition of sensitive scenes by adding adversarial frame to the image. Experimental results indicate that our method can achieve strong adversarial effects while maintaining the visual appeal and social function of the transformed images.

Original languageEnglish
JournalCEUR Workshop Proceedings
Volume2670
StatePublished - 2019
Externally publishedYes
Event2019 Working Notes of the MediaEval Workshop, MediaEval 2019 - Sophia Antipolis, France
Duration: 27 Oct 201930 Oct 2019

Fingerprint

Dive into the research topics of 'Adversarial photo frame: Concealing sensitive scene information of social images in a user-acceptable manner'. Together they form a unique fingerprint.

Cite this