Sven Hellbach, Julian Eggert, Edgar Körner, Horst-Michael Groß, "Basis Decomposition of Motion Trajectories Using Spatio-temporal NMF", Artificial Neural Networks - {ICANN} 2009, 19th International Conference, pp. 804–814, 2009.

This paper’s intention is to present a new approach for decomposing motion trajectories. The proposed algorithm is based on nonnegative matrix factorization, which is applied to a grid like representation of the trajectories. From a set of training samples a number of basis primitives is generated. These basis primitives are applied to reconstruct an observed trajectory. Hence, the reconstruction information can be used for classification. Furthermore, the basis primitives can be used to predict the observed movement. For the experiments, real movement data is used to evaluate several aspects of the proposed approach. In particular, the focus of the experiments is set to the prediction task.