Augmented reality systems are ones which insert virtual content into a user’s view of the real world, in response to environmental conditions and the user’s behavior within that environment. This virtual content can take the form of visual elements such as 2D labels or 3D models, auditory cues, or even haptics; content is generated and updated based on user behavior and environmental conditions, such as the user’s location, movement patterns, and the results of computer vision or machine learning operations. AR systems are used to solve problems in a range of domains, from tourism and retail, education and healthcare, to industry and entertainment. For example, apps from Lowe’s [82] and Houzz [81] support retail transactions by scanning a user’s environment and placing product models into the space, thus allowing the user to preview what the product might look like in her home. AR systems have also proven helpful in such areas as aiding industrial assembly tasks [155, 175], helping users overcome phobias [35], and reviving interest in cultural heritage sites [163].
Mobile AR systems are ones which run on portable handheld or wearable devices, such that the user is free to move around their environment without restric- tion. Examples of such devices include smartphones, tablets, and head-mounted dis- plays. This freedom of movement and usage, in combination with the application’s reliance on computer vision and machine learning logic to provide core function- ality, make mobile AR applications very difficult to test. In addition, as demand and prevalence of machine learning logic increases, the availability and power of commercially available third-party vision libraries introduces new and easy ways for developers to violate usability and end-user privacy.
The goal of this dissertation, therefore, is to understand and mitigate the challenges involved in testing mobile AR systems, given the capabilities of today’s commercially available vision and machine learning libraries. We consider three related challenge areas: application behavior during unconstrained usage conditions, general usability, and end-user privacy. To address these challenge areas, we present three research efforts. The first presents a framework for collecting application performance and usability data in the wild. The second explores how commercial vision libraries can be exploited to conduct machine learning operations without user knowledge. The third presents a framework for leveraging the environment itself to enforce privacy and access control policies for mobile AR applications. / Computer and Information Science
| Identifer | oai:union.ndltd.org:TEMPLE/oai:scholarshare.temple.edu:20.500.12613/7775 |
| Date | January 2022 |
| Creators | Lehman, Sarah, 0000-0002-9466-0688 |
| Contributors | Tan, Chiu C., Ling, Haibin, Bai, Li, Payton, Jamie |
| Publisher | Temple University. Libraries |
| Source Sets | Temple University |
| Language | English |
| Detected Language | English |
| Type | Thesis/Dissertation, Text |
| Format | 170 pages |
| Rights | IN COPYRIGHT- This Rights Statement can be used for an Item that is in copyright. Using this statement implies that the organization making this Item available has determined that the Item is in copyright and either is the rights-holder, has obtained permission from the rights-holder(s) to make their Work(s) available, or makes the Item available under an exception or limitation to copyright (including Fair Use) that entitles it to make the Item available., http://rightsstatements.org/vocab/InC/1.0/ |
| Relation | http://dx.doi.org/10.34944/dspace/7747, Theses and Dissertations |
Page generated in 0.0018 seconds