Deep-cARe System: Projection-based Augmented Reality with Deep Learning for Elderly Healthcare in Indoor

Deep-cARe System: Projection-based Augmented Reality with Deep Learning for Elderly Healthcare in Indoor

As societies age, the development of advanced and ubiquitous intelligent technologies that provide medical support and enhance the health of the elderly has become increasingly important. In their everyday lives, elderly people also encounter episodes of pain, including the loss of cognitive and memory skills. We propose a projection-based augmented reality (PAR) device fitted with a deep-learning module to provide auxiliary functions and ensure the safety of the elderly in everyday living situations. We suggest a three-dimensional space reconstruction of an omnipresent PAR space for the elderly in this review. To lay the foundation for contextual knowledge, we propose the application of a deep-learning module. Performance tests were carried out on the PAR technology to graft the deep learning structure (pose estimation, face recognition, and object detection) through the proposed hardware to check the probability of execution, real-time execution, and applicability. Furthermore, according to the scenarios, some user interfaces (UIs) were introduced regardless of the distance between users and the proposed system. In this research, we built a bidirectional PAR system that provides the relevant information instead of a unidirectional PAR system for simple information provision by understanding the user environment and action intentions. By combining PAR and deep-learning architectures, we present a discussion of the possibility of care systems for the elderly.

Author(s) Details

Yoon Jung Park
Media System Lab, Yonsei University, South Korea.

Hyochoel Ro
Media System Lab, Yonsei University, South Korea.

Nam Kyu Lee
College of Computer Science & Software, Chung-Ang University, South Korea.

Tack-Don Han
Media System Lab, Yonsei University, South Korea.

View Book :- https://bp.bookpi.org/index.php/bpi/catalog/book/320

Editor 251News

leave a comment

Create Account



Log In Your Account