They are all after you

Abstract

Many of the authentication schemes for mobile devices that were proposed lately complicate shoulder surfing by splitting the attacker's attention into two or more entities. For example, multimodal authentication schemes such as Gaze-TouchPIN and GazeTouchPass require attackers to observe the user's gaze input and the touch input performed on the phone's screen. These schemes have always been evaluated against single observers, while multiple observers could potentially attack these schemes with greater ease, since each of them can focus exclusively on one part of the password. In this work, we study the effectiveness of a novel threat model against authentication schemes that split the attacker's attention. As a case study, we report on a security evaluation of two state of the art authentication schemes in the case of a team of two observers. Our results show that although multiple observers perform better against these schemes than single observers, multimodal schemes are significantly more secure against multiple observers compared to schemes that employ a single modality. We discuss how this threat model impacts the design of authentication schemes.

RepliCueAuth: Validating the Use of a Lab-Based Virtual Reality Setup for Evaluating Authentication Systems

“…In both the replication study and the real-world study [52] the attacks were single-person attacks through optimal views on the authentication scheme and the interaction -this depicts a best case scenario for the attacker. Using non-optimal or user-defined views, or more advanced threat models (e.g., multiple observers [48]), could result in different findings. Third, following Khamis et al 's [52] study design means facing the same limitations.…”

The Role of Eye Gaze in Security and Privacy Applications: Survey and Future HCI Research Directions

“…Suitable threat models should be employed when evaluating the security of authentication schemes. For example, previous work evaluated multimodal authentication schemes against two observers [76] and by using video record-ings from two cameras [74]. Commonly studied threats include shoulder surfing attacks [39], video attacks [32], thermal attacks [1], and smudge attacks [11].…”

CueAuth

“…However, we expect that performing it against gaze input would still be challenging due to the large number of cues that the attacker would have to observe, in addition to the eye movements. Another direction for future work is to evaluate our methods against other observation-based threat models, such as the insider model [60], and the case of having multiple attackers observation the user simultaneously [31]. The latter is particularly relevant for public displays, where it is typical that a group surrounds the user during interaction [32].…”