![]() ![]() Although previous work has investigated an operating system enforced AR output security module which relies on developer written policies, we previously realized that, while promising, these hand-coded policies can be difficult to define for real-world use. Sim- ilarly, a hologram which displays the speedometer should not be obstructed by a hologram which displays the album art of the song the driver is currently playing. For example, in the case of displaying holograms in car windshields, it would be dangerous for a hologram to obstruct a stop sign. Preventing holograms with a lower priority from obstructing holograms with a higher priority.Regulating visual content displayed to reduce distraction and prevent obstruction of the real-world context.$15.00 In particular, visual output security is concerned with two issues pertaining to the user’s visual field : SenSys ’19, November 10–13, 2019, New York, NY, USA © 2019 Copyright held by the owner/author(s). For all other uses, contact the owner/author(s). Copyrights for third-party components of this work must be honored. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. AR security is concerned with both the inputs and outputs of AR devices. While the increasing proliferation of AR devices will undoubtedly enable many new applications, issues of privacy and security cannot be ignored. Re- search has shown that AR will be a $100 billion industry by 2020 and companies are actively developing AR technologies for consumer and commercial use. ![]() 1 INTRODUCTION Augmented Reality (AR) is becoming increasingly ubiquitous. In The 17th ACM Conference on Embedded Networked Sensor Systems (SenSys ’19), November 10–13, 2019, New York, NY, USA. Demo Abstract: Adaptive AR Visual Output Security using Reinforcement Learning Trained Policies. ACM Reference Format: Joseph DeChicchis, Surin Ahn, and Maria Gorlatova. ![]() KEYWORDS Augmented reality, visual output security, reinforcement learning, policy optimization, Magic Leap AR headset. Computing meth- odologies → Mixed / augmented reality Reinforcement learn- ing.Security and privacy → Systems security.The demonstration illustrates that RL based visual output security systems are feasible. We develop a visual output security application using a RL trained policy and deploy it on a Magic Leap One head- mounted AR device. However, whether such systems and policies can be deployed on a physical AR device without degrading performance was left an open question. Previous work has proposed techniques for securing the output of AR devices and used reinforcement learning (RL) to train security policies which can be difficult to define manu- ally. New security challenges arise as AR becomes increasingly ubiquitous. Demo Abstract: Adaptive AR Visual Output Security using Reinforcement Learning Trained Policies Joseph DeChicchis Duke University Durham, NC, USA Surin Ahn Stanford University Palo Alto, CA, USA Maria Gorlatova Duke University Durham, NC, USA ABSTRACT Augmented reality (AR) technologies have seen significant improve- ment in recent years with several consumer and commercial solu- tions being developed.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |