Visually-imparied People's Perception of Camera-based Technologies

HCI Research


Paper title: "If sighted people know, I should be able to know:" Privacy Perceptions of Bystanders with Visual Impairments around Camera-based Technology.

Link: www.usenix.org

Conference: Paper Submitted to USENIX Security Conference 2023

Author: Yaxing Yao, Yuhang Zhao, Jiaru Fu, Nihan Zhou



Overview

I worked in the HCI Accessibility lab and was the main researcher of the study in understanding visually impaired people’s (VIP’s) perception of sensing technology especially camera-based devices around them.


01

Preparation process

At the beginning of the research process, I performed the literature review in privacy, accessibility, camera technology usage with people with disabilities, and relevant HCI qualitative research methods. I was able to summarize and analyze the ideas established in the papers, found surprising points and gaps that could be further explored and addressed in our research. By reading prior works, I had the opportunity to identify needs for further research and problems we could address in our project, for example, the perception of camera use of visual-impaired people as bystanders.

Furthermore, I worked collaboratively with Professor Zhao to categorize the previous researches and understand our project goals. I had a chance to communicate closely with Professor Zhao and another colleague in defining our potential participants, setting up the project structure, timeline, and methods, establishing factors and aspects for analysis. I also received a license from UW Social and Behavioral CITI training program for doing human-centered research and drew guidelines and storyboards of how we can approach our research topic and work with visually impaired people to obtain useful data.



02

Research

Survey

I participated in defining and creating scenario-based survey questions and generating a Qualtric survey. I reached out to multiple organizations to help us spread the surveys as well as utilizing social media such as Reddit and Facebook to recruit participants. After collecting all the survey results, I utilized Python to perform statistical analysis and testing in the qualitative data and was finally able to present graphical visualization of the result.

Interview

I was able to organize the survey results to find potential participants for further in-depth interviews. I conducted interviews with participants who were either low visions or blind. During the interview process and by communicating with my participants, I received useful data for further analysis.

After the interview sessions, I create a content-coding thematic analysis for qualitative data by employing affinity diagrams. I was able to generate code, themes, and categories for our interview responses that helped me analyze the data in a more structured way.



03

Result

With all the data collection and analysis, I have done, I was able to organize and find patterns in these results and collaborate with my professor and other researchers in writing a structured paper report in Overleaf, and finally submit the paper to the USENIX security conference.