Mechanisms of nonverbal communication: Emotion recognition based on facial expressions and analysis of head and body postures (C03)

在子项目C3中，开发了人机交互过程中非语言信号的视觉检测和处理的新方法和新技术。重点分析了精细结构的微表情作为不可隐藏的身体状态参数的形式、弱表达的手势以及多人场景中的人体姿势估计。尤其令人感兴趣的是，当使用移动传感器时，可以在多大程度上使用这些方法。此外，还开发了处理基于事件的传感器稀疏编码数据的神经形态算法。这些方法为移动系统领域的计算机视觉任务提供了新的解决方案。

In subproject C3 new methods and technologies for the visual detection and processing of non-verbal signals in human-machine-interaction processes are developed. They focus on the analysis of fine-structured micro-expressions as forms of non - concealable body state parameters, weakly expressed gestures and the human pose estimation in multi-person scenarios. It is of particular interest to which extent these methods can be employed when mobile sensors are used. Moreover, neuromorphic algorithms for processing sparse coded data of event-based sensors are developed. These approaches provide new solutions to computer vision tasks in the field of mobile systems.