Visible to the public Protecting Multimedia Privacy from Both Humans and AI

TitleProtecting Multimedia Privacy from Both Humans and AI
Publication TypeConference Paper
Year of Publication2019
AuthorsLiu, Bo, Xiong, Jian, Wu, Yiyan, Ding, Ming, Wu, Cynthia M.
Conference Name2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)
Keywordsadversarial image perturbation, AI, AI-assisted attackers, AI-assisted malicious entities, artificial intelligence, data privacy, Deep Learning, face recognition, human adversaries, Human Behavior, human factors, image, Image coding, image privacy, multimedia, multimedia data, multimedia privacy issues, privacy, privacy protection, private information, protecting multimedia privacy, pubcrawl, resilience, Resiliency, Scalability, security of data
AbstractWith the development of artificial intelligence (AI), multimedia privacy issues have become more challenging than ever. AI-assisted malicious entities can steal private information from multimedia data more easily than humans. Traditional multimedia privacy protection only considers the situation when humans are the adversaries, therefore they are ineffective against AI-assisted attackers. In this paper, we develop a new framework and new algorithms that can protect image privacy from both humans and AI. We combine the idea of adversarial image perturbation which is effective against AI and the obfuscation technique for human adversaries. Experiments show that our proposed methods work well for all types of attackers.
Citation Keyliu_protecting_2019