Continuous Body and Hand Gesture Recognition for Natural Human-Computer Interaction
Name
SongDD2011c.pdf
Description
Accepted version
Size
6.68 MB
Format
Adobe PDF
Checksum (MD5)
4c0c732f0ba4044be40a0eac5cef8a3e
Author(s)
Song, Yale
Demirdjian, David
Davis, Randall
Date Issued
2010
Citation
Song, Yale, Demirdjian, David and Davis, Randall. 2010. "Continuous Body and Hand Gesture Recognition for Natural Human-Computer Interaction."
Version
Author's final manuscript
Abstract
We present a new approach to gesture recognition that tracks body and hands simultaneously and recognizes gestures continuously from an unseg-mented and unbounded input stream. Our system estimates 3D coordinates of upper body joints and classifies the appearance of hands into a set of canonical shapes. A novel multi-layered filtering technique with a temporal sliding window is developed to enable online sequence labeling and segmentation. Experimental results on the NATOPS dataset show the effectiveness of the approach. We also report on our recent work on multimodal gesture recognition and deep-hierarchical sequence representation learning that achieve the state-of-the-art performances on several real-world datasets.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
