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

Figure 12

The model’s accuracy and the model’s time for animal images as input in 10 runs of the feedback mode and feedback-less mode for inverted images in three levels of categorization. (a) The model’s accuracy of dog and cat’s inverted images. (b) The model’s time of dog and cat’s inverted images. (c) The model’s accuracy of dolphin and fish’s inverted images. (d) The model’s time of dolphin and fish’s inverted images. (e) The model’s accuracy of spider and butterfly’s inverted images. (f) The model’s time of spider and butterfly’s inverted images.

(a)

(b)

(c)

(d)

(e)

(f)