@inproceedings{ba3a3e5ddf174c62b38d099bbe588806,
title = "3D human activity recognition using skeletal data from RGBD sensors",
abstract = "In this paper, a new effective method was proposed to recognize human actions based on RGBD data sensed by a depth camera, namely Microsoft Kinect. Skeleton data extracted from depth images was utilized to generate 10 direction features which represent specific body parts and 11 position features which represent specific human joints. The fusion features composed of both was used to represent a human posture. An algorithm based on the difference level of adjacent postures was presented to select the key postures from an action. Finally, the action features, composed of the key postures{\textquoteright} features, were classified and recognized by a multiclass Support Vector Machine. Our major contributions are proposing a new framework to recognize the users{\textquoteright} actions and a simple and effective method to select the key postures. The recognition results in the KARD dataset and the Florence 3D Action dataset show that our approach significantly outperforms the compared methods.",
author = "Jiaxu Ling and Lihua Tian and Chen Li",
note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2016.; 12th International Symposium on Visual Computing, ISVC 2016 ; Conference date: 12-12-2016 Through 14-12-2016",
year = "2016",
doi = "10.1007/978-3-319-50832-0\_14",
language = "英语",
isbn = "9783319508313",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "133--142",
editor = "George Bebis and Darko Koracin and Tobias Isenberg and Sandra Skaff and Amela Sadagic and Richard Boyle and Fatih Porikli and Jianyuan Min and Carlos Scheidegger and Alireza Entezari and Bahram Parvin and Daisuke Iwai",
booktitle = "Advances in Visual Computing - 12th International Symposium, ISVC 2016, Proceedings",
}