Investigating the intelligibility of a computer vision system for blind users

Abstract

Computer vision systems to help blind users are becoming increasingly common yet often these systems are not intelligible. Our work investigates the intelligibility of a wearable computer vision system to help blind users locate and identify people in their vicinity. Providing a continuous stream of information, this system allows us to explore intelligibility through interaction and instructions, going beyond studies of intelligibility that focus on explaining a decision a computer vision system might make. In a study with 13 blind users, we explored whether varying instructions (either basic or enhanced) about how the system worked would change blind users' experience of the system. We found offering a more detailed set of instructions did not affect how successful users were using the system nor their perceived workload. We did, however, find evidence of significant differences in what they knew about the system, and they employed different, and potentially more effective, use strategies. Our findings have important implications for researchers and designers of computer vision systems for blind users, as well more general implications for understanding what it means to make interactive computer vision systems intelligible. CCS CONCEPTS• Human-centered computing~Human computer interaction (HCI) • Human-centered computing~Accessibility • Computing methodologies~Computer vision

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

“…Three participants (P4, P8, P10) mentioned having light perception, and four participants reported having experience with wearable cameras such as Aira [11] with frequencies shown in Table 1. Responses to technology attitude questions 5 indicate that they are positive about technology in general (𝑚𝑒𝑎𝑛 = 2.02, 𝑆𝐷 = 0.42) and wearable technology such as smartglasses (𝑚𝑒𝑎𝑛 = 1.89, 𝑆𝐷 = 0.90).…”

“…Twenty-four self-identified as male and fifteen as female, and one chose not to disclose their gender. Participants' responses to technology attitude questions 5 indicate that they were slightly positive about wearable technology (𝑚𝑒𝑎𝑛 = 1.08, 𝑆𝐷 = 0.95).…”

Shared Privacy Concerns of the Visually Impaired and Sighted Bystanders with Camera-Based Assistive Technologies

“…One concern could be the use of an online survey, which gives a verbal description of a more abstract technology rather than a paratyping [21] or a prototype-based [18,86] study. These approaches have their merits but have the limitation that the speciics of the artifact may bias the participant and take the focus away from the actual information that is being accessed and shared.…”

Disability-first Dataset Creation: Lessons from Constructing a Dataset for Teachable Object Recognition with Blind and Low Vision Data Collectors

“…There are currently some efforts to make AI transparent to a wider audience [34,40], as well as educate end-users about AI [17, 36,52]. However, there is little research to investigate how to make AI for accessibility transparent to people with disabilities [3] and to our knowledge there are no curricula for teaching about AI for accessibility to the disabled community. Your project will need to consider how to overcome these challenges and educate your target community.…”