Analytical review on human activity recognition in video

The main motive of this review paper is to recognise the human activities in video using different posses and various types of activities done by human in video. To achieve this activity recognition author's used a different technique such as object segmentation, feature extraction and representation, Hidden markov model, bag of word approach. And some basic concepts of machine learning and algorithms such as supervised learning, clustering, Linear Discriminant analysis, Finite state automata, K-Nearest Neighbour have been used. The domain area for this analysis is surveillances, entertainment and healthcare environment. And the authors have collected the data for their analysis from various sources such as Youtube, movies, real human activities videos are collected from Railway stations, banks, hospitals, circus area specially which are under the camera notification.