Human Activity Recognition using RGB-D Sensors

Computer Vision (CV) is a very alive area of research. The CV has all around multiple applications. One application of CV Ambient Assisted Living (AAL). AAL is getting very much importance day by day because of its extraordinary social values. In this paper we were considering AAL we focused on human activity recognition for that we are using knight depth camera for capturing 3-D skeleton data. This paper is about human activity recognition by using knight depth camera. This camera will capture 3-dimensional skeleton data.[1] Human Activity Recognition deals with a bag of multiple frame sequences with the image caption. Many methodologies have been proposed previously. In this paper, Convolutional Neural Network is used which gives the accuracy 0.85 on SBU Kinect Interaction Dataset.