Video surveillance refers to the automated monitoring and analysis of visual data captured by cameras in public or controlled environments, with the goal of understanding crowd behavior, detecting anomalies, and ensuring public safety. In the context of WiFi/CSI and crowd sensing research, it matters because it establishes the benchmark problem domain — tasks such as crowd counting, density estimation, and activity recognition — against which sensing-based approaches are often compared or proposed as privacy-preserving alternatives. Key variants include static camera surveillance focused on single-image analysis, video-based sequential monitoring that leverages temporal information, and multi-camera systems designed to handle occlusion and large-scale scene coverage.

Source Papers

  • A survey of recent advances in CNN-based single image crowd counting and density estimation — A survey of recent advances in CNN-based single image crowd
  • Recent trends in crowd analysis: A review — Recent trends in crowd analysis: A review