Synthèse de mouvements humains par des méthodes basées apprentissage: un état de l’art

Due to the increasing number of graphical applications, the generation of human life-like characters has become an important research topic. Different approaches have been proposed, the combination of motion capture data and machine learning methods being the dominant trend in the last years. Despite the good results produced by these approaches, to our knowledge there does not exist a document surveying the learning-based methods that intend to solve the problem of re-using and generalizing motion capture data. In this article, we present a state-of-the-art of the most recurrent methods used for the synthesis of whole-body kinematic motions. We present the principles and ideas behind each one of them, as well as the type of representations and pre-processing steps applied over motion capture data before learning.