Emotion Sensing with Wearable Devices

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

Final Year Thesis Oral Defense

Title: "Emotion Sensing with Wearable Devices"

by

CHIU Mang Tik

Abstract:

Building systems that have the ability to recognize human emotions has 
attracted much interest in recent years. Common approaches toward machine 
emotion recognition focus on detection of facial expressions and analysis 
of physiological signals. However, in situations where these features 
cannot be easily obtained, emotion recognition becomes a challenging 
problem. In this thesis, we explore the possibility of emotion recognition 
through gait, which is one of the most common human behaviors. We take two 
main approaches towards emotion inference, one with statistical feature 
extraction and selection, and the other with motion time-series analysis. 
In both approaches, motion features are obtained by human pose estimation 
from captured video frames. These features are then either extracted based 
on predefined statistical features for machine learning model training, or 
are used to compute movement differences in the time domain. Performance 
comparison shows the best performing model achieved 74% accuracy on 
classifying five emotion labels. Finally, we implement a proof-of-concept 
wearable-server system for emotion recognition in real-life scenarios 
using smart glass cameras.


Date            : 23 April 2018 (Monday)

Time            : 18:00 - 18:40

Venue           : Room 2127B (via lift 19), HKUST

Advisor         : Dr. HUI Pan

2nd Reader      : Prof. KWOK James Tin-Yau