In Computer Vision and image processing field, understanding the actions of human from videos is demanding task. Humans performing the actions in the video is automatically cognized and designating their actions is the prime purpose of astute video systems. Its motive is to acknowledge the actions and objectives of one or more multiple objects from a series of examination on the action of objects and their environmental condition. Human action recognition has gained popularity because of its wide applicability varying from Content-based Video Analytics, Visual Surveillance and Video Indexing etc. This paper depicts survey on person to person interaction, single person and multiple people action recognition and deciphers the methodologies and limitations of diverse methods for human action recognition are explored.