3D upper body pose estimation is a topic greatly studied by the computer vision society because it is useful in a great number of applications, mainly for human robots interactions including communications with companion robots. However there is a challenging problem: the complexity of classical algorithms that increases exponentially with the dimension of the vectors’ state becomes too difficult to handle. To tackle this problem, we propose a new approach that combines several annealing particle filters defined independently for each limb and belief propagation method to add geometrical constraints between individual filters. Experimental results on a real human gestures sequence will show that this combined approach leads to reliable results.

中文翻译：

---

结合退火粒子滤波和信念传播进行3D上身跟踪

3D上半身姿势估计是计算机视觉协会研究的主题，因为它在许多应用程序中很有用，主要用于人机交互（包括与伴侣机器人的通信）。然而，存在一个具有挑战性的问题：经典算法的复杂性随着向量状态的维数成倍增加而变得难以处理。为了解决这个问题，我们提出了一种新方法，该方法结合了为每个肢体独立定义的几个退火粒子过滤器和信念传播方法，以在各个过滤器之间添加几何约束。在真实的人类手势序列上的实验结果将表明，这种组合方法可提供可靠的结果。