Communications on Applied Electronics |
Foundation of Computer Science (FCS), NY, USA |
Volume 4 - Number 4 |
Year of Publication: 2016 |
Authors: Deepak N.A., Sinha U.N. |
10.5120/cae2016652080 |
Deepak N.A., Sinha U.N. . Analysis of Human Gait for Person Identification and Human Action Recognition. Communications on Applied Electronics. 4, 4 ( February 2016), 1-4. DOI=10.5120/cae2016652080
The gait sequence analysis finds applications in several fields such as, person identification, human action recognition and event classification. The extraction of gait sequences is the initial step in gait sequence analysis. The Latent Dirichlet Allocation (LDA) a type of ’Topic Model’ is used to analyse the gait sequences for person identification and human action recognition. This is achieved by proposing a novel method that transforms the gait sequences into words representation. Finally, the generated words are then analysed using LDA for person identification and human action recognition. The proposed person identification algorithm is tested using CASIA gait dataset A and human action recognition algorithm is tested using Weizmann action recognition dataset, resulting in correct classification rate of 85%using CASIA Dataset A and 85% using Weizmann action recognition dataset.