In this work, we select three public video datasets commonly used for violence recognition and crowd behavior understanding: Real-Life Violence Situations (RLVS), RWF–2000, and HockeyFight. These datasets include aggressive behaviors such as interactions with crowds, crowd provocation, and non-violent actions in difficult, uncontrolled real-world settings.

For this research, we employ three publicly available video datasets that are frequently utilized in violence detection and crowd behavior analysis: Real-Life Violence Situations (RLVS), RWF-2000, and HockeyFight. These datasets encompass various forms of aggressive conduct, including crowd interactions, provocative behaviors, and non-violent activities captured within challenging, uncontrolled real-world environments.