Poster: Video Chat Scam Detection Leveraging Screen Light Reflection

摘要: The rapid advancement of social media and communication technology enables video chat to become an important convenient way daily communication. However, such convenience also makes personal clips easily obtained exploited by malicious users who launch scam attacks. Existing studies only deal with the attacks that use fabricated facial masks, while liveness detection targets playback using a virtual camera is still elusive. In this work, we develop novel system, which can track weak light changes reflected off skin human face leveraging chromatic eigenspace differences. We design inconspicuous challenge frame minimal intervention robust anomaly detector verify remote user in session. Furthermore, propose resilient defense strategy defeat both naive intelligent spatial temporal verification. evaluation results show our system achieve accurate accuracy false rate as high 97.7% (94.8%) 1% (1.6%) on smartphones (laptops), respectively.