A review on Human Action Recognition in videos using Deep Learning (original) (raw)

Human Action Recognition (HAR) in video plays a vital role in today's world. The aim of HARis to automatically identify and analyse human activities using acquired information from video data. Some of the applications include security and surveillance, smart homes and assisted living, health monitoring, robotics, human– computer interaction, intelligent driving, video-retrieval, gaming and entertainment etc. This paper explores the impact of Deep Learning techniques on action recognition. We also explore how spatiotemporal features are aggregated through various deep architectures, the role of optical flow as an input, the impacts on real-time capabilities, and the compactness & interpretability of the learned features. Although several papers have already been published in the general HAR topics, the growing technologies in the field as well as the multi-disciplinary nature of HAR prompt the need for constant updates in the field. In this respect, this paper attempts to review ...