Document Type

Technical Report

Department

Computer Science and Engineering

Publication Date

2012

Filename

WUCSE-2012-36.pdf

DOI:

10.7936/K7154F8R

Technical Report Number

WUCSE-2012-36

Abstract

he goal of this research project was to be able to identify and track a hand using the depth image from a Microsoft Kinect. The ability to do this would have uses in sign language recognition, rehabilitation, and gesture recognition amongst others. The Microsoft Kinect is currently capable of identifying and tracking whole bodies. Our approach was to follow the method that Microsoft used but apply it to detailed hand recognition instead of the body. The basic strategy they used was to take a large amount of labeled depth data and build a decision tree to identify which part of the body each pixel of an image belongs to.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7154F8R

Download

Share

COinS