Inflated 3D Convolution Network for Detecting Anomalies in Surveillance Videos

Abstract

All cities are getting smart with the intervention of latest technologies, there infrastructure is getting upgraded with each day. Critical informations is provided to us by these infrastructures.With the rise in popularity of AI, there is a requirement for a real-time system that can aid in spotting crimes as they occur.The surveillance platform’s information may include both aberrant and conventional footage. We propose developing an aberrant event identification system based on weakly annotated training videos, and so when such behaviour is discovered, suitable action may be taken.For extraction of features, we deployed I3D-Resnet-50, a deep deep residual model. The Kinetics video action dataset was used to train this network. There are 13 unique abnormalities in our dataset. Crime, Attack, Firing, Burglaries, Thieving, Prison, Fight, Thefts, Breaking and entering, Bomb, Criminal damage, Torture, and Traffic Accident are all unusual incidents. The proposed approach for visual anomaly detection achieves considerable improvements in terms of correctness and recall.