State-of-the-arts Violence Detection using ConvNets
Aayush Jain, Dinesh Kumar Vishwakarma
Abstract
Over the last few years, there has been a surge in developing automated systems that can monitor human activities in the public environment and recognize abnormalities like violent and suspicious events. Violence detection aims at identifying whether a violent action has occurred and has evolved as a popular topic in the field of image processing and computer vision. Effective techniques for accurate and intelligent video analysis are highly demanded. There are various methodologies which can detect such activities. This paper aims at reviewing various CNN based state-of-the-art techniques for violence detection. In this paper, we discuss variants of CNN, their application and datasets which play an indispensable part in the detection process.