The use of depth sensors in activity recognition is a technology that emerges in human computer interaction and motion recognition. In this study, an approach to identify single-person activities using deep learning on depth image sequences is presented. First, a 3D volumetric template is generated using skeletal information obtained from a depth video. The generated 3D volume is used for extracting features by taking images from different angles at different volumes. Actions are recognized by extracting deep features using AlexNet model  and Histogram of Oriented Gradients (HOG) features from these images. The proposed method has been tested with MSRAction3D  and UTHKinect-Action3D  datasets. The obtained results were comparable to similar studies in the literature.