Advertisement
Published in:
2018 | OriginalPaper | Chapter
Human Action Recognition with Skeleton Data Using Extreme Learning Machine
Authors : Ying Li, Xiong Luo, Weiping Wang, Wenbing Zhao
Published in: Proceedings of 2017 Chinese Intelligent Automation Conference
Publisher: Springer Singapore
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
Skeleton-based human action recognition has recently drawn many researchers’ attention with the prevalence of Microsoft Kinect sensors. In this paper, a method of action recognition using the skeletal features is proposed to improve the classification performance and reduce the training time to a large extent. We first select several key frames in each action sequence which can represent the corresponding action significantly, and extract meaningful joint-based and body part-based features of key frames. Then, we use extreme learning machine (ELM) algorithm to achieve action recognition. We compare the proposed approach with other state-of-the-art methods on a large public dataset to evaluate the performance. The experimental results indicate that our approach can achieve a good recognition performance with extremely fast training speed, while applying the proposed method to the online action recognition system.