Live Video Synopsis for Multiple Cameras

TL;DR

This paper introduces a live video synopsis system for multi-camera surveillance that displays past activities from Slave cameras when people appear in the Master camera, aiding real-time decision making.

Contribution

It proposes an action tube-based method for live video synopsis that handles multiple people, activities, and re-identification uncertainties in a hierarchical camera setup.

Findings

Effective in displaying relevant past activities in real-time

Handles multiple objects and activities simultaneously

Improves decision-making in live surveillance

Abstract

Video surveillance cameras generate most of recorded video, and there is far more recorded video than operators can watch. Much progress has recently been made using summarization of recorded video, but such techniques do not have much impact on live video surveillance. We assume a camera hierarchy where a Master camera observes the decision-critical region, and one or more Slave cameras observe regions where past activity is important for making the current decision. We propose that when people appear in the live Master camera, the Slave cameras will display their past activities, and the operator could use past information for real-time decision making. The basic units of our method are action tubes, representing objects and their trajectories over time. Our object-based method has advantages over frame based methods, as it can handle multiple people, multiple activities for each person, and can address re-identification uncertainty.