Tracking the activities of daily lives: An integrated approach

摘要

The tracking of the activities of daily living (ADL) may have significant implications in healthcare because it would enable healthcare professionals to receive updates remotely regarding the functional status of post-injury and post-surgery patients, and people with disabilities and the elderly. If successful, technologies that enable ADL tracking could dramatically reduce the healthcare cost because people could stay at the comfort of their home instead of at the healthcare facilities. In this paper, we present the design and implementation of a system that integrates two modalities of human motion tracking: computer vision and inertial sensing. For the computer vision modality, we choose to use the Microsoft Kinect sensor due to its low-cost and excellent programming support. For the inertial sensing modality, we choose to use smart watches for the same reason. The Kinect sensor is responsible for indoor ADL tracking, while the smart-watch component is responsible for outdoor ADL tracking. The smart-watch component also facilitates user-identification with the Kinect sensor and user's health related activities information with indoor computing component, as well delivers realtime feedbacks to the users.