Title

Real time academic emotion recognition using body gestures

Date of Publication

2012

Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Suarez Merlin

Abstract/Summary

Emotion plays a powerful role in students learning process, however, many intelligent tutoring systems (ITS) only consider the knowledge models of the students. Hence, building affect models for ITSs have been an emerging research researchers had proposed several approaches on affect recognition based on the students facial expressions, speech and body gestures. However, most of these works focused on gathering emotional information from facial expressions and speech of the students. Acted emotion databases and expensive hardware are also used to predict student emotions. In this paper, a markerless approach is proposed to detect four basic academic emotions (boredom, flow, confusion, frustration) from the student based on their body gestures. The data for this study were gathered from five students using Microsoft Kinect camera. Features extracted were hand position and duration, arm position, speed, head tilt, leaning and shifting. Given the features extracted, the gestures detected by the system include hands up/hands near face/hands down, arms up/arms down, scratch/steady, fast/slow, lean forward/lean backward, lean right/lean left, shifting/not shifting. To classify the emotion of the student, rules were generated based on the students gestures. The system generated five rules for each students, the system achieved an accuracy of 72.57%, 91.06% , 76.42% , 77.91% and 81.96% for each student respectively.

Abstract Format

html

Language

English

Format

Print

Accession Number

TG05246

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

1 v. (various foliations) ; 28 cm. + 1 computer optical disc.

Keywords

Emotion recognition

This document is currently not available here.

Share

COinS