Accessing Passersby Proxemic Signals through a Head-Worn Camera: Opportunities and Limitations for the Blind

Abstract

Fig. 1. A pedestrian detection scenario demonstrating the data collection in our study. A blind person, wearing smart glasses with our working prototype, and a sighted person walk toward each other in a corridor. The smart glasses detect the sighted passerby, estimate his proximity, and share his relative location and head pose.The spatial behavior of passersby can be critical to blind individuals to initiate interactions, preserve personal space, or practice social distancing during a pandemic. Among other use cases, wearable cameras employing computer vision can be used to extract proxemic signals of others and thus increase access to the spatial behavior of passersby for blind people. Analyzing data collected in a study with blind (N=10) and sighted (N=40) participants, we explore: (i) visual information on approaching passersby captured by a head-worn camera; (ii) pedestrian detection algorithms for extracting proxemic signals such as passerby presence, relative position, distance, and head pose; and (iii) opportunities and limitations of using wearable cameras for helping blind people access proxemics related to nearby people. Our observations and findings provide insights into dyadic behaviors for assistive pedestrian detection and lead to implications for the design of future head-worn cameras and interactions. CCS Concepts: • Human-centered computing → User studies; Empirical studies in HCI; Mobile devices; Empirical studies in accessibility; Accessibility technologies; • Computing methodologies → Computer vision.

From the Lab to People's Home: Lessons from Accessing Blind Participants' Interactions via Smart Glasses in Remote Studies

“…Our analysis of video frames from Vizux Blade, which has a 64-degree horizontal field of view, indicates a higher percentage of visibility from the glasses for participants who became blind later in life compared to those who were born blind and tend to "turn their head to orient their ear towards a sound source" [31,33,51]. This observation provides additional evidence on prior work indicating how limited field of view in smart glasses for pedestrian detection can exclude blind users who could benefit the most from this technology [27]. While smart glasses with a wider camera angle could potentially help, they can also lead to image distortions [48].…”

“…We believe that cameras embedded in smart glasses can play a role in filling this gap as they can provide a mobile and first-person perspective, which may or may not capture both the interface and participant's interactions with it. To our knowledge, using smart glasses for this task has not been previously explored with blind participants, whose idiosyncratic head movements could potentially affect what is being captured by the camera [27].…”

“…Assuming that the field of view of the camera worn by users with visual impairments can capture the area of interest to them, all the prior work proposes assistive systems that interpret the visual input and convert it into non-visual formats, such as audio and haptic. While many focused on assistive applications of smart glasses for people with visual impairments, Lee et al focused on blind people's camera aiming behaviors in the context of pedestrian detection [27]. They observed that smart glasses worn by blind people tend to capture a passerby well in a corridor but may exclude a passerby from the camera frame when they are close to the passerby due to the behavior of focusing on the sound source (i.e., the passerby's voice in their case).…”

“…Prior to leveraging smart glasses in remote studies, two sighted researchers (R1 and R5) had extensive experience with employing this technology for in-person studies with blind participants. Specifically, in collaboration with other sighted and blind collaborators, they used a pair of Vuzix Blade smart glasses [50] to explore blind people's camera aiming behaviors for pedestrian detection [27] and understand its social acceptability [28].…”

Challenges in Acceptance of Smartphone-Based Assistive Technologies: Extending the UTAUT Model for People With Visual Impairments

“…Another factor affecting acceptance of assistive technology, according to Lee et al (2020), is users’ perception of technological solutions, which, in turn, depends on whether users have direct or indirect experience with them. The present study highlights the significant positive effect of the former on users’ opinions about assistive technology.…”