Figure 14.

Comparison between our classifier, CNN-1 [25] and CNN-2 [35] based on the patch-based medium RAFT-MIM${}^{Q_{12}}$ dataset.