Style of action based individual recognition in video sequences

We present a method for recognizing individuals from their ldquostyle of actionrdquo. Two forms of human recognition can be useful: the determination that an object is from the class of humans (which is called human detection), and the determination that an object is a particular individual from this class (this is called individual recognition). This paper focuses on the latter problem. A periodicity is detected in from a sequence of motion detected binary image frames by finding the maximum similarity measure between them. Based on the periodicity information the Motion History Image (MHI) is applied for each individual sequence to find out entire motion information of periodic action. The individual is then recognized using a partial Hausdorff Distance similarity measure and the SVM classification approach.