Posts by Collection

publications

Automatically Detecting Bystanders in Photos to Reduce Privacy Risks

IEEE Symposium on Security and Privacy, 2020

People often appear as other (usually unknown) people’s photos as bystanders, and when these photos are shared online, they pose great privacy threats towards them, especially during an era when advances in machine learning enables adversaries to automatically search, identify, and track people utilizing huge image databases available in the cloud. We propose a machine learning model to automatically detect bystanders in an image, so that they can be obfuscated before before posting that photo online. Read more

A SOCIO-TECHNICAL APPROACH TO PROTECTING PEOPLE’S PRIVACY IN THE CONTEXT OF SHARING IMAGES ON SOCIAL MEDIA

Indiana University, 2020

My dissertation explores a socio-technical approach to protect our privacy in the context of sharing images online, when we are not in full control over sharing our visual data. It proposes machine learning-based technical solutions to detect and obfuscate sensitive image contents, as well as behavioral interventions to encourage social media users to respect and protect others’ privacy. Read more