Date of Award
May 2024
Degree Type
Thesis
Degree Name
Master of Science
Department
Computer Science
First Advisor
Rohit J Kate
Committee Members
Scott Strath, Zeyun Yu
Keywords
3D-CNN, Energy Expenditure, Frame Differences, Machine Learning, Optical Flow, Physical Activity
Abstract
This thesis presents a novel approach for predicting energy expenditure of physical activity from videos using optical flows and deep learning. Conventional approaches mainly rely on wearable sensors, which, despite being widely used, are constrained by practicality and accuracy concerns. This proposal introduces a new strategy that utilizes a three-dimensional Convolutional Neural Network (3D-CNN) to evaluate video data and accurately estimate energy costs in metabolic equivalents (METs). Our model utilizes optical flow extraction to analyze video, capturing complex motion patterns and their changes over time. The results are good indicating potential for this method to be deployed in various healthcare applications, such as automatic health monitoring and physical activity surveillance. This research contributes towards accurate automatic estimation of energy expenditure of physical activity simply from recorded videos and thus creates opportunities for non-invasive health monitoring systems.
Recommended Citation
Kasturi, Gayatri, "PREDICTING ENERGY EXPENDITURE FROM PHYSICAL ACTIVITY VIDEOS USING OPTICAL FLOWS AND DEEP LEARNING" (2024). Theses and Dissertations. 3483.
https://dc.uwm.edu/etd/3483