Spatial-Channel Attention-Based Class Activation Mapping for Interpreting CNN-Based Image Classification Models

<div>Class activation maps generated by SCA-CAM, GAP-CAM, and Grad-CAM under four networks. Networks from the first row to the fourth row are SqueezeNet, ResNet-18, ResNet-50, and DenseNet-161, respectively.</div>

Security and Communication Networks

fig12

Figure 12

Figure 12: Spatial-Channel Attention-Based Class Activation Mapping for Interpreting CNN-Based Image Classification Models 