A Recurrent Temporal Model for Semantic Levels Categorization Based on Human Visual System

Figure 7

The potential of neurons categorization, for different objects at three levels of classification, is shown over time. These categories are separated from each other in a way that the classifier would be able to recognize them alongside of the time-increasing. There are 10, 5, and, 2 categories in the subordinate, basic, and superordinate levels, respectively. The shapes around the graph depict the separation of the categories at times 100, 700, 3000, and 5500 using a k-means classifier, and the notable point is that the assumed time is the model time, which ranges from 0 to 7225.