作者：Laura Leal-Taix´e,Gerard Pons-Moll and Bodo Rosenhahn

作者单位：Institute for Information Processing (TNT)

发布时间：2012

发布期刊/会议：CVPR

论文全称：Branch-and-price global optimization for multi-view multi-target tracking

论文地址：

论文代码：

地位：

个人理解

• **创新点：**在都将多视图多目标跟踪任务分为reconstruction和tracking两个子任务的背景下，本文提出了一种新的多视图多目标跟踪的全局优化公式，将多目标多视图跟踪任务当做一个全局优化问题（将该问题转化为图结构问题），并引入了分解和分支方法来解决该问题
• **为什么：**将多视图多目标跟踪任务分为reconstruction和tracking两个子任务的算法效果不好，且比较耗时。
• 怎么做：在本文中，作者没有把多摄像头多目标跟踪任务分为两阶段问题来解决，而是提出了一个范式在单独的全局优化问题中执行2D-3D分配（重建）和时间分配（跟踪），将多视图多目标跟踪任务转换为图结构问题，图的边是成本，即最短路径问题。接着将设计的图结构问题转化为一个线性程序，最后使用Branch-and-price for multi-commodity flow方法进行求解。

一、摘要

We present a new algorithm to jointly track multiple objects in multi-view images. While this has been typically addressed separately in the past, we tackle the problem as a single global optimization. We formulate this assignment problem as a min-cost problem by defifining a graph structure that captures both temporal correlations between objects as well as spatial correlations enforced by the confifiguration of the cameras. This leads to a complex combinatorial optimization problem that we solve using Dantzig-Wolfe decomposition and branching. Our formulation allows us to solve the problem of reconstruction and tracking in a single step by taking all available evidence into account. In several experiments on multiple people tracking and 3D human pose tracking, we show our method outperforms state-of-the-art approaches.

• 本文提出了一种在多视图图像中联合跟踪多个目标的新算法。
• 虽然这在过去通常被分开单独处理，但我们将这个问题作为一个单独的全局优化问题来处理。我们通过定义一个图结构，既捕获物体之间的时间相关性，也捕获由摄像机配置强制执行的空间相关性，从而将这个分配问题表述为一个最小成本问题，这导致了一个复杂的组合优化问题，作者使用 Dantzig-Wolfe 分解和分支来解决。