ABSTRACT
Anomaly detection in surveillance videos plays a important role in ensuring public safety and security in various scenarios such as transportation hubs, public events, and urban areas. This paper comprehensively examines current techniques for detection of anomaly in surveillance videos. This paper also provides an overview of the hurdles associated with anomaly detection in scenes, discuss the evolution of techniques from traditional methods to deep learning-based approaches, and analyze the strengths and limitations of each approach. Additionally, this paper also has a proposed supervised model that works on avenue dataset to find the anomaly with 91% of accuracy.
