Analysis and performance of three methods of human action recognition
Rajeev Shrivastava, Amit Pratap Singh · 2012
Abstract— Human action recognition is one of the most important emerging trend /technology .It has wide application such as surveillance (behavior analysis), security (pedestrian detection), control (human-computer interfaces), content based video retrieval, etc. There are many methods of human action recognition .The human action recognition problem is made difficult by the great variability in body part rotation and tilt, lighting intensity and angle, body part movement , aging, partial occlusion (e.g. Wearing Hats, scarves, glasses etc.), etc.Principal components from the body parts space are used for human action recognition to reduce dimensionality. A multiscale representation human action recognition is done to preserve the discriminate information prior to dimensionality reduction. In this paper, three multiscale representation techniques Gabor filter; Log Gabor filter and Discrete Wavelet Transform are applied prior to dimensionality reduction. PCA is then applied on the above techniques to find the human recognition accuracy rate and to compare the results of the three methods with PCA method. In this paper we have compared efficiency of three methods of human action recognition namely Gabor, Log Gabor and discrete wavelet transform.