Skip To Main Content
Support Knowledge Base

How to Generate a Heatmap Using OpenVINO™?

Content Type: Troubleshooting   |   Article ID: 000056835   |   Last Reviewed: 11/09/2023

Environment

OpenVINO C++ PackageCNN

Description

Can a heatmap generation technique like layer-wise relevance propagation or Grad-CAM be implemented using OpenVINO™?

Resolution

Activation maps can be used to visualize the decision-making.

In order to calculate Class Activation Maps in OpenVINO™, is necessary to:

  • Access the output feature maps of the last layer and the corresponding weights.
    NoteThis can be done through the CNNNetwork: addOutput() function call when initializing the network.

After running the inference, calculate the weighted sum of the weights with the feature maps and use this to generate the visualization.

Related Products

This article applies to 1 products.