作者：Xindi Zhang，Ebroul Izquierdo

作者单位：Multimedia and Vision Research Group, Queen Mary University of London, London, UK

发布时间：2019

发布期刊/会议：IEEE Visual Communications and Image Processing (VCIP)

论文全称：Real-Time Multi-Target Multi-Camera Tracking with Spatial-Temporal Information

论文地址：

Real-Time Multi-Target Multi-Camera Tracking with Spatial-Temporal Information

论文代码：

地位：

个人理解

• 创新点：
  • 开发了一种多目标多摄像头实时在线跟踪行人的算法
  • 给跟踪目标添加了新的跟踪状态（有4种跟踪状态：active，lost，searching，remove）
• 为什么：
  • 智能视频监控识别嫌疑人是计算机视觉领域的一个活跃的研究领域，而已有的性能较好的在线多目标跟踪算法基本都是基于单摄像头的，而性能较好的多摄像头多目标跟踪算法基本都是离线的，结果的输出有延迟，不适合用于智能视频监控系统，故本文提出一个在线的实时的多摄像头多目标跟踪算法。
• 怎么做：
  • 使用Openpose[3]实现人物检测。然后将Openpose提取的关键点转移到检测框坐标中
  • 将单摄像头跟踪器MOTDT[4]扩展为多摄像头跟踪器
  • 根据摄像头之间的位置关系，总结了可能的相机变化的时间间隔分布的直方图

一、摘要

Video security monitoring has always been an important mission for safety reason. In an entire surveillance system, there are usually several cameras distributed sparsely to cover a wide range of public areas (e.g., school, shopping mall or infrastructure). Tracking person through this cameras network is challenging due to different camera perspectives, illumination changes and pose variations. Several algorithms for Multi-Target Multi-Camera tracking (MTMCT) have been proposed in offlfline method which has delay in getting result. Addressing the need for real-time computation of people tracks through multi-camera, the paper proposes an online tracking algorithm. The contributions include (1) online real-time framework which can be used in practical application, (2) extend a single camera multi object tracking (MOT) algorithm to be suitable for multi-camera tracking and (3) use spatial-temporal information to strengthen cross camera person recall performance. The proposed algorithm has been benchmarked against the literature review of MTMC algorithms.

• 由于不同的摄像头视角，照明变化和姿势变化，跟踪跨越多个摄像头的目标是具有挑战性的
• 目前已有的多摄像头多目标跟踪算法基本都是离线算法，存在延时的问题