jinglescode/papers

Interpretable and Fine-Grained Visual Explanations for Convolutional Neural Networks

jinglescode opened this issue · 1 comments

Paper

Title: Interpretable and Fine-Grained Visual Explanations for Convolutional Neural Networks
Authors: Jorg Wagner, Jan Mathias Kohler, Tobias Gindele, Leon Hetzel, Jakob Thaddaus Wiedemer, Sven Behnke
Link: https://openaccess.thecvf.com/content_CVPR_2019/html/Wagner_Interpretable_and_Fine-Grained_Visual_Explanations_for_Convolutional_Neural_Networks_CVPR_2019_paper.html
Year: 2019

Summary

  • produce mask to focus on interpretability
  • smallest region of image must be retained to preserve (or deleted to change) model output
  • fine grain visual explanation, no smoothing and regularisations

Hi, I'm very interested in your work. I want to try out this method for my thesis about Decoding the Art of Robot Tactile Learning with Explainable Neural Networks for Incipient Slip Sensing. Would you be willing to share the code of FGVis?