People de-identification plays a major role in addressing legal and privacy concerns in video-based services. In private spaces, where there may be only one person present (e.g. an older person living alone), user identification might not be an issue, as they would have approved the disclosure of data to service providers or other stakeholders. Concerns are more related to the disclosure of appearance (e.g. if the person is dressed/naked) and behaviour. In public spaces, issues arise with the identification of those persons interacting with the user or by-standers, who may not have agreed to be recorded. Therefore, a privacy-by-design approach would need to be followed in the development of video-based AAL systems. However, ensuring privacy may have special influence in the possibility of providing some services. For instance, if an application stores images in order to allow people with mild dementia to recall events or persons, de-identifying those persons (e.g. by blurring their faces) will have a negative effect in the service provided. Hence, a trade-off would need to be achieved between privacy protection and utility of the data. Research projects Research Project 13 - Privacy preservation in video-based AAL applications Research Project 14 - Context recognition for the application of visual privacy