Human-robot-interaction

Unsupervised autoencoder learns spatial context from trajectory data for annotation.