GradCAM

Grad-CAM is a popular technique for visualizing the parts of an image that a Convolutional Neural Network (CNN) is looking.

Gradient-weighted Class Activation Mapping (Grad-CAM) is a class discriminative localization tool in classification tasks. It uses the gradients of any target concept (say ‘dog’ in a classification network or a sequence of words in a captioning network) flowing into the final convolutional layer to produce a coarse localization map highlighting the important regions in the image for predicting the concept. In this task, Grad-CAM is used to localize the snake’s position and pattern in the image.