To main content

Principal Feature Visualisation in Convolutional Neural Networks

Abstract

We introduce a new visualisation technique for CNNs called Principal Feature Visualisation (PFV). It uses a single forward pass of the original network to map principal features from the final convolutional layer to the original image space as RGB channels. By working on a batch of images we can extract contrasting features, not just the most dominant ones with respect to the classification. This allows us to differentiate between several features in one image in an unsupervised manner. This enables us to assess the feasibility of transfer learning and to debug a pre-trained classifier by localising misleading or missing features.

Category

Academic article

Client

  • Research Council of Norway (RCN) / 259869

Language

English

Author(s)

Affiliation

  • Norges miljø- og biovitenskapelige universitet
  • SINTEF Digital / Smart Sensor Systems

Year

2020

Published in

Lecture Notes in Computer Science (LNCS)

ISSN

0302-9743

Publisher

Springer

Volume

12368

Page(s)

18 - 31

View this publication at Cristin