Visual Hand Tracking and Gesture Analysis

This dissertation presents novel techniques to effectively and efficiently track articulate hand motion in a video sequence. Since tracking articulate hand motion is equivalent to a search in a high-dimensional space, an analysis of the feasible configuration space is given, and semiparametric and nonparametric representations are proposed to model the motion constraint. For each of the representations, a corresponding efficient tracking algorithm is proposed. Experiments have shown the effectiveness of these approaches. To fully automate the tracking system, this dissertation also presents an algorithm for automatic initialization which can be done in real time. Based on this result, several vision-based gesture interfaces are built.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Lin, John

Contributors dc:contributor

• Huang, Thomas S.

• Engineering, Electronics and Electrical

Language dc:language

eng

Identifier

(MiAaPQ)AAI3160919

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/80894