Searching of Gender from Biological Motion using Matlab

Abstract

Biological motion are those which contains information about the identity of an agent as well as about his or her actions, intentions, and emotions etc. It also contains many form of information. Observers are usually able to tell ‘what’ action is being performed (e.g.: walking versus running),’how’ it is being performed (e.g.: quickly versus slowly) and by ‘whom’ (e.g.: a young versus an old actor).The human visual system is highly sensitive to biological motion and is capable to extract socially relevant information from it. Here we investigate the question of ‘How such information is encoding in biological motion patterns and how such information can be easily retrieved. A framework is developed that transforms biological motion into a visual representation allowing for analysis using linear methods from statistical patterns .By Considering Classification of gender as an instance, we can say simple classifiers are constructed and compared to psycho-physical data from human observers. Here, we use the visual search to explore perception of gender-from-motion. Observers were asked to point a figure showing walking of a male or female gate among distractors having the same form but opposite motion. The point-light walkers moved along random path in a 3-D visual environment. The set size is varied between 1 and 4, and targets were presents on 50% of the trials. The proposed framework can be used not only for biological motion analysis but also to arrange new motion patterns.