Violent action recognition has significant importance in developing automated video surveillance systems. Over the last few years, violence detection such as fight activity recognition is mostly been achieved through hand-crafted features detectors. Some researchers also inquired about learning-based representation models. These approaches achieved high accuracies on Hockey and Movies benchmark datasets specifically designed for the detection of violent sequences. However, these techniques have limitations in learning discriminating features for videos with abrupt camera motion of the Hockey dataset. Deep representation-based approaches have been successfully used in image recognition and human action detection tasks. This research work project used a deep representation-based model using the concept of transfer learning for violent scenes detection to identify aggressive human behaviors. The result reports that the proposed approach is outperforming state-of-the-art accuracies by learning the most discriminating features achieving 99.28% and 99.97% accuracies on Fight activity videos such as Hockey and Movies datasets respectively, by learning the finest features for the task of violent action recognition in videos.
SlimLoigx 2020. All Rights Reserved