Older adults' privacy considerations for vision based recognition methods of eldercare applications

This study aims to explore older adults' privacy considerations for technology based monitoring applications in eldercare that use video systems. It specifically aims to introduce alternative vision based tools and identify whether distorting or "anonymizing" captured images affect older adults' privacy concerns and willingness to allow such an application to be installed in their residence. Ten residents of an independent retirement community were recruited to participate in a series of scenarios. Each scenario involved a daily activity such as sitting in the living room and having a visitor, or preparing a snack. These sessions were video-recorded using different image processing and extraction approaches. Follow-up in-depth interviews with participants were conducted after a demonstration of the captured images. Findings indicate that shape extraction can alleviate privacy concerns associated with the use of cameras. Participants expressed no privacy concerns with silhouette images and emphasized the importance of anonymity in the video sequences. They furthermore expressed the desire to control system operation by being able to turn a vision-based system off and on, and also determine who has access to the collected information.