Fig. 5From: COGNIMUSE: a multimodal video database annotated with saliency, events, semantics and emotion with application to summarizationThe two-dimensional space of arousal-valenceBack to article page