Relevant LMA Features for Human Motion Recognition

International Journal of Information Technology

Insaf Ajili ; Malik Mallem ; Jean-Yves Didier

关键词: Human motion recognition;    Discriminative LMA features;    random forest;    features reduction.;   

DOI  :  10.1999/1307-6892/10009574

学科分类：计算机应用

来源: World Academy of Science, Engineering and Technology (W A S E T)

【 摘 要 】

Motion recognition from videos is actually a very complex task due to the high variability of motions. This paper describes the challenges of human motion recognition, especially motion representation step with relevant features. Our descriptor vector is inspired from Laban Movement Analysis method. We propose discriminative features using the Random Forest algorithm in order to remove redundant features and make learning algorithms operate faster and more effectively. We validate our method on MSRC-12 and UTKinect datasets.

【 授权许可】

Unknown