With increasing applications such as human-computer interaction, surveillance, medical diagnosis and computer animation, automatic understanding and modelling of human motion have attracted more attention from researchers and industries. Extensive research in human motions especially facial behavior and human body actions has been explored in recent years. The bulk of research for human motion understanding concentrates on using low-level machine-understanding features. Despite the huge amount of excellent research in these fields, the effective and efficient description and representation of human motions remain challenging in many scenarios. Therefore, the aim of this special issue is to survey the-state-of-the-art methodologies, algorithms and concepts in advanced human motion understanding with any kind of data types. It intends to bridge the gap between the low-level features and high-level semantics of human motion.