University of South Florida
Multimodal Assessment of Human Behavior with Applications in Analysis of Autism Spectrum Disorder
Abstract
dc:descriptionHuman behavior analysis is an important application area of artificial intelligence. Usingdata collected from sensors (e.g., cameras), human behavior including, but not limited to, emotions, expressions, body movement, and eye gaze are analyzed. This analysis is useful in applications in diverse fields ranging from healthcare, defense, education, and marketing. Considering this, in this dissertation, we have proposed new approaches to emotion and expression recognition, subject identification, and classification of autism spectrum disorder. More specifically, this dissertation has the following contributions: 1) A new approach for multimodal ubiquitous emotion recognition is proposed; 2) Investigation into the impact of large expression variations on 3D subject identification is detailed; 3) An approach to generalizing expression across age ranges (children vs. elderly) is proposed; 4) This is the first work to propose using self-report questionnaire data to classify autism spectrum disorder across age ranges; and (5) This is the first work to proposed a context-based approach to classifying videos of children at risk for autism spectrum disorder.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy (Ph.D.)
- Level thesis:degree_level
- Dissertation
- Discipline thesis:degree_discipline
- Computer Science and Engineering
- Grantor
- University of South Florida
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jannat, Sk Rahatul
- Contributors dc:contributor
-
- Shaun Canavan, Ph.D.
- Dmitry Goldgof, Ph.D.
- Heather Agazzi, Ph.D.
- Tempestt Neal, Ph.D.
- Pei-Sung Lin, Ph.D.
Subjects
dc:subject × 5Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.usf.edu/etd/10760