Computer Science Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/2756

Browse

Now showing 1 - 4 of 4

Accessible On-Body Interaction for People With Visual Impairments
(2016) Oh, Uran Oh; Findlater, Leah; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
While mobile devices offer new opportunities to gain independence in everyday activities for people with disabilities, modern touchscreen-based interfaces can present accessibility challenges for low vision and blind users. Even with state-of-the-art screenreaders, it can be difficult or time-consuming to select specific items without visual feedback. The smooth surface of the touchscreen provides little tactile feedback compared to physical button-based phones. Furthermore, in a mobile context, hand-held devices present additional accessibility issues when both of the users’ hands are not available for interaction (e.g., on hand may be holding a cane or a dog leash). To improve mobile accessibility for people with visual impairments, I investigate on-body interaction, which employs the user’s own skin surface as the input space. On-body interaction may offer an alternative or complementary means of mobile interaction for people with visual impairments by enabling non-visual interaction with extra tactile and proprioceptive feedback compared to a touchscreen. In addition, on-body input may free users’ hands and offer efficient interaction as it can eliminate the need to pull out or hold the device. Despite this potential, little work has investigated the accessibility of on-body interaction for people with visual impairments. Thus, I begin by identifying needs and preferences of accessible on-body interaction. From there, I evaluate user performance in target acquisition and shape drawing tasks on the hand compared to on a touchscreen. Building on these studies, I focus on the design, implementation, and evaluation of an accessible on-body interaction system for visually impaired users. The contributions of this dissertation are: (1) identification of perceived advantages and limitations of on-body input compared to a touchscreen phone, (2) empirical evidence of the performance benefits of on-body input over touchscreen input in terms of speed and accuracy, (3) implementation and evaluation of an on-body gesture recognizer using finger- and wrist-mounted sensors, and (4) design implications for accessible non-visual on-body interaction for people with visual impairments.
Exploring Blind and Sighted Users’ Interactions With Error-Prone Speech and Image Recognition
(2021) Hong, Jonggi; Kacorri, Hernisa; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Speech and image recognition, already employed in many mainstream and assistive applications, hold great promise for increasing independence and improving the quality of life for people with visual impairments. However, their error-prone nature combined with challenges in visually inspecting errors can hold back their use for more independent living. This thesis explores blind users’ challenges and strategies in handling speech and image recognition errors through non-visual interactions looking at both perspectives: that of an end-user interacting with already trained and deployed models such as automatic speech recognizer and image recognizers but also that of an end-user who is empowered to attune the model to their idiosyncratic characteristics such as teachable image recognizers. To better contextualize the findings and account for human factors beyond visual impairments, user studies also involve sighted participants on a parallel thread. More specifically, Part I of this thesis explores blind and sighted participants' experience with speech recognition errors through audio-only interactions. Here, the recognition result from a pre-trained model is not being displayed; instead, it is played back through text-to-speech. Through carefully engineered speech dictation tasks in both crowdsourcing and controlled-lab settings, this part investigates the percentage and type of errors that users miss, their strategies in identifying errors, as well as potential manipulations of the synthesized speech that may help users better identify the errors. Part II investigates blind and sighted participants' experience with image recognition errors. Here, we consider both pre-trained image recognition models and those fine-tuned by the users. Through carefully engineered questions and tasks in both crowdsourcing and semi-controlled remote lab settings, this part investigates the percentage and type of errors that users miss, their strategies in identifying errors, as well as potential interfaces for accessing training examples that may help users better avoid prediction errors when fine-tuning models for personalization.
Interactive Sonification of Abstract Data - Framework, Design Space, Evaluation, and User Tool
(2006-04-24) Zhao, Haixia; Shneiderman, Ben; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
For people with visual impairments, sound is an important information channel. The traditional accommodation for visually impaired users to access data is to rely on screen readers to speak the data in tabular forms. While speech can accurately describe information, such data presentation tends to be long and hard to realize complex information. This is particularly true in exploratory data analysis in which users often need to examine the data from different aspects. Sonification, the use of non-speech sound, has shown to help data comprehension. Previous data sonifications focus on data to sound attribute mapping and typically lack support for task-oriented data interaction. This dissertation makes four contributions. (1) An Action-by-Design-Component (ADC) framework guides auditory interface designs for exploratory data analysis. The framework characterizes data interaction in the auditory mode as a set of Auditory Information Seeking Actions (AISA). It also discusses design considerations for a set of Design Components to support AISAs, contrasted with actions in visualizations. (2) Applying the framework to geo-referenced statistical data, I explore its design space. Through user evaluations, effective design options were identified and insights were obtained regarding human ability to perceive complex information, especially those with spatial structures, from interactive sounds. (3) A tool, iSonic, was developed, with synchronized visual and auditory displays. Forty-two hours of case studies with seven blind users show that iSonic enables them to effectively explore data in highly coordinated map and table views without special devices, to find facts and discover data trends even in unfamiliar geographical contexts. Preliminary algorithms are also described to automatically generate geographical region spatial sweep orders for arbitrary maps. (4) The application to geo-referenced data demonstrated that the ADC framework provided a rich set of task-oriented actions (AISAs) that were effective for blind users to accomplish complex tasks with multiple highly coordinated data views. It also showed that some widely used techniques in visualization can adapt to the auditory mode. By applying the framework to scatterplots and line graphs, I show that the framework could be generalized and lead to the design of a unified auditory workspace for general exploratory data analysis.
Ubiquitous Accessibility Digital-Maps for Smart Cities: Principles and Realization
(2019) Ismail, Heba; Agrawala, Ashok; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
To support disabled individuals' active participation in the society, the Americans with Disabilities Act (ADA) requires installing various accessibility measures in roads and public accommodation spaces such as malls and airports. For example, curb ramps are installed on sidewalks to aid wheel-chaired individuals to transition from/to sidewalks smoothly. However, to comply with the ADA requirements, it is sufficient to have one accessible route in a place and usually there are no clear directions on how to reach that route. Hence, even within ADA-compliant facilities, accessing them can still be challenging for a disabled individual. To improve the spaces' accessibility, recently, systems have been proposed to rate outdoor walkways and intersections’ accessibility through active crowdsourcing where individuals mark and/or validate a maps’ accessibility assessments. Yet, depending on humans limits the ubiquity, accuracy and the update-rate of the generated maps. In this dissertation, we propose the AccessMap—Accessibility Digital Maps—system to build ubiquitous accessibility digital-maps automatically; where indoor/outdoor spaces are updated with various accessibility semantics and marked with assessment of their accessibility levels for the vision- and mobility-impairment disability types. To build the maps automatically, we propose a passive crowdsourcing approach where the users’ smartphone devices’ spatiotemporal sensors signals (e.g. barometer, accelerometer, etc.) are analyzed to detect and map the accessibility semantics. We present algorithms to passively detect various semantics such as accessible pedestrian signals and missing curb-ramps. We also present a probabilistic framework to construct the map while taking the uncertainty in the detected semantics and the sensors into account. AccessMap was evaluated in two different countries, the evaluation results show high detection accuracy for the different accessibility semantics. Moreover, the crowdsourcing framework helps further improve the map integrity overtime. Additionally, to tag the crowdsourced data with location stamps, GPS is the de-facto-standard localization method, but it fails in indoor environments. Thus, we present the Hapi WiFi-based localization system to estimate the crowdsourcers’ location indoors. WiFi represents a promising technology for indoor localization due to its world-wide deployment. Nevertheless, current systems either rely on a tedious expensive offline calibration phase and/or focus on a single-floor area of interest. To address these limitations, Hapi combines signal-processing, deep-learning and probabilistic models to estimate a user’s 2.5D location (i.e. the user floor-level and her 2D location within that floor) in a calibration-free manner. Our evaluation results show that, in high-rise buildings, we could achieve significant improvements over state-of-the-art indoor-localization systems.