Ambient Intelligence Empowered Smart Nursing Home for Vision-Based Elderly Caring

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Defense

Title: "Ambient Intelligence Empowered Smart Nursing Home for Vision-Based 
Elderly Caring"

by

LI Jiaxiu

Abstract:

Recent advances in computer vision offer a viable pathway to alleviate the 
growing demand for human labor in elderly care. Among the core challenges in 
smart caregiving, fall risk remains a critical concern for older adults. 
While existing approaches achieve satisfactory accuracy in post hoc fall 
detection, comparatively little attention has been devoted to anticipatory 
identification of fall risk prior to incident occurrence. To bridge this gap, 
we formalise fall risk detection as a predictive task and propose a new 
training framework that enables early risk assessment. The project 
demonstrates a model with improved performance relative to existing methods 
while maintaining lower operational cost, highlighting its potential for 
scalable and proactive elderly care applications.

Date            : 28 April 2026 (Tuesday)

Time            : 16:50 - 17:30

Venue           : Room 2132A (near Lift 19), HKUST

Advisor         : Dr. CHEN Hao

2nd Reader      : Prof. CHAN Gary Shueng-Han