Privacy Concerns for Visual Assistance Technologies

Stangl, Abigale; Davis, Nathan; Xie, Bin; Fleischmann, Kenneth R.; Findlater, Leah; Gurari, Danna

doi:10.1145/3517384

Cited by 16 publications

(3 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…P1 reflected on the lack of concerns as “we either got numb or gave up.” Such attitudes could be explained by their nuanced understanding of data policies. For example, camera-based assistive technologies like Aira or SeeingAI which our participants reported to use, provide no clear indication of whether and for what purpose personal visual data are shared with third parties [ 130 ]. This perhaps could lead to something of a paradox widely discussed in privacy literature [ 78 ].…”

Section: Resultsmentioning

confidence: 99%

“…Brown [ 29 ] found that, despite the general worries people seemed to have about privacy, they would still give out information for perceived benefits. In fact, the two participants who reacted to both scenarios ( i.e ., potential risk cases of re-identification, data abuse and misuse) with stronger concerns and hesitancy to data sharing, reported to use Be My Eyes—one of the few services that explicitly indicates dissemination of video streams to third parties [ 130 ].…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Contributing to Accessibility Datasets: Reflections on Sharing Study Data by Blind People

Kamikubo

Lee

Kacorri

2023

Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

To ensure that AI-infused systems work for disabled people, we need to bring accessibility datasets sourced from this community in the development lifecycle. However, there are many ethical and privacy concerns limiting greater data inclusion, making such datasets not readily available. We present a pair of studies where 13 blind participants engage in data capturing activities and reflect with and without probing on various factors that influence their decision to share their data via an AI dataset. We see how different factors influence blind participants' willingness to share study data as they assess risk-benefit tradeoffs. The majority support sharing of their data to improve technology but also express concerns over commercial use, associated metadata, and the lack of transparency about the impact of their data. These insights have implications for the development of responsible practices for stewarding accessibility datasets, and can contribute to broader discussions in this area. CCS CONCEPTS• Human-centered computing → Human computer interaction (HCI); Accessibility; • Social and professional topics → People with disabilities; • Security and privacy → Human and societal aspects of security and privacy.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Contributing to Accessibility Datasets: Reflections on Sharing Study Data by Blind People

Kamikubo

Lee

Kacorri

2023

Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

show abstract

“…To effectively train AI models for these varied use cases, it is crucial to create comprehensive datasets that reflect a wide range of scenarios, information display preferences, and output modalities. If these datasets will contain sensitive information pertaining to BLV people -for example, to capture the scenario of watching videos from friends and family -we must also recognize and consider BLV users' visual privacy concerns [46,112,113,137].…”

Section: Applying Generative Ai To Explore the Video Accessibility De...mentioning

confidence: 99%

“It’s Kind of Context Dependent”: Understanding Blind and Low Vision People’s Video Accessibility Preferences Across Viewing Scenarios

Jiang,

Jung,

Phutane

et al. 2024

Proceedings of the CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

While audio description (AD) is the standard approach for making videos accessible to blind and low vision (BLV) people, existing AD guidelines do not consider BLV users' varied preferences across viewing scenarios. These scenarios range from how-to videos on YouTube, where users seek to learn new skills, to historical dramas on Netflix, where a user's goal is entertainment. Additionally, the increase in video watching on mobile devices provides an opportunity to integrate nonverbal output modalities (e.g., audio cues, tactile elements, and visual enhancements). Through a formative survey and 15 semi-structured interviews, we identified BLV people's video accessibility preferences across diverse scenarios. For example, participants valued action and equipment details for howto videos, tactile graphics for learning scenarios, and 3D models for fantastical content. We define a six-dimensional video accessibility design space to guide future innovation and discuss how to move from "one-size-fits-all" paradigms to scenario-specific approaches. CCS CONCEPTS• Human-centered computing → Accessibility.

show abstract

Analysis and design framework for the development of indoor scene understanding assistive solutions for the person with visual impairment/blindness

Valipoor,

de Antonio,

Cabrera

2024

Multimedia Systems

View full text Add to dashboard Cite

This paper discusses the challenges of the current state of computer vision-based indoor scene understanding assistive solutions for the person with visual impairment (P-VI)/blindness. It focuses on two main issues: the lack of user-centered approach in the development process and the lack of guidelines for the selection of appropriate technologies. First, it discusses the needs of users of an assistive solution through state-of-the-art analysis based on a previous systematic review of literature and commercial products and on semi-structured user interviews. Then it proposes an analysis and design framework to address these needs. Our paper presents a set of structured use cases that help to visualize and categorize the diverse real-world challenges faced by the P-VI/blindness in indoor settings, including scene description, object finding, color detection, obstacle avoidance and text reading across different contexts. Next, it details the functional and non-functional requirements to be fulfilled by indoor scene understanding assistive solutions and provides a reference architecture that helps to map the needs into solutions, identifying the components that are necessary to cover the different use cases and respond to the requirements. To further guide the development of the architecture components, the paper offers insights into various available technologies like depth cameras, object detection, segmentation algorithms and optical character recognition (OCR), to enable an informed selection of the most suitable technologies for the development of specific assistive solutions, based on aspects like effectiveness, price and computational cost. In conclusion, by systematically analyzing user needs and providing guidelines for technology selection, this research contributes to the development of more personalized and practical assistive solutions tailored to the unique challenges faced by the P-VI/blindness.

show abstract

Privacy Concerns for Visual Assistance Technologies

Abstract: BoulderCCS Concepts: • Human-centered computing → Empirical studies in accessibility.

Cited by 16 publications

References 76 publications

Contributing to Accessibility Datasets: Reflections on Sharing Study Data by Blind People

Contributing to Accessibility Datasets: Reflections on Sharing Study Data by Blind People

“It’s Kind of Context Dependent”: Understanding Blind and Low Vision People’s Video Accessibility Preferences Across Viewing Scenarios

Analysis and design framework for the development of indoor scene understanding assistive solutions for the person with visual impairment/blindness

Contact Info

Product

Resources

About