Demo of PassFrame: Generating image-based passwords from egocentric videos

We demonstrate a personalized user authentication mechanism based on first-person-view videos. Our proposed algorithm forms temporary image-based authentication challenges which benefit a variety of purposes such as unlocking a mobile device or fallback authentication. First, representative frames are extracted from the egocentric videos. Then, they are split into distinguishable segments before repetitive scenes are discarded through a clustering procedure. We integrate eye tracking data to select informative sequences of video frames and suggest an alternative method based on image quality. For evaluation, we perform experiments in different settings including object-interaction activities and traveling contexts. We assessed the authentication scheme in the presence of an informed attacker and observed that the entry time is significantly higher than that of the legitimate user.