PrivacyTag: A Community-Based Method for Protecting Privacy of Photographed Subjects in Online Social Networks

Abstract

Online social networks, such as Facebook, have become popular with people of all ages, and online communication with friends and acquaintances via messages that include photos has become very common. With the increasing ease with which users can take and post photos, the unintentional disclosure of sensitive information of various kinds through mistakes made while posting has become a problem. In this work, we focused on the privacy of people appearing in photos and developed a method called "PrivacyTag" for adaptively blurring their facial area in accordance with the communities to which they belong by using tags embedded with community-based privacy policies. We also evaluated a newly designed privacy tag and developed a prototype application for Facebook that uses this tag.