Evaluation of Skeleton Trackers and Gesture Recognition for Human-robot Interaction.
Translated title
Evaluering af menneske- og gestikulerings-genkendelse til menneske-robot interaktion.
Author
Term
4. term
Publication year
2013
Submitted on
2013-06-05
Pages
83
Abstract
This report is a master thesis in the field Vision, Graphics and Interactive System. It documents the work done during a two semesters visit at Georgia Institute of Technology. The report consists of two parts: An evaluation of skeleton trackers and Gesture recognition. The evaluation of skeleton trackers compares the two Kinect skeleton trackers from Microsoft and Primesense. The results of the tests shows that their performance is very similar. It is also tested how a simple multi Kinect setup with data fusion can work, which shows slight improvements. The gesture recognition is based on hidden Markov models and is mainly about recognizing pointing gestures. The test includes a two-class and a multi-class setup, which both yields good results.
Keywords
Kinect ; Microsoft ; OpenNI ; Primesense ; NITE ; gestikulering
Documents