a review on video violence detection approaches

A violent behaviour detection system (vbds) is an important application of intelligent video surveillance that performs a critical role in the field of public security and safety vbds is a sort of behaviour recognition that seeks to determine whether the behaviours observed in the situation are violent, such as fighting or assault. this paper presents a survey of the existing approaches to vbds. in this paper, the existing vbds techniques are classified based on their framework, which includes the old-fashion framework and the end-to-end state-of-the-art deep learning framework. finally, the vbds methods’ performance is assessed and compared.