Video Summarization Using a Neurodynamical Model of Visual Attention