Principal Feature Visualisation in Convolutional Neural Networks

Autor:	A. Berge, Marianne Bakken, Johannes Kvam, Alexey Stepanov
Rok vydání:	2020
Předmět:	Computer science Feature (computer vision) business.industry Principal (computer security) Principal component analysis RGB color model Pattern recognition Artificial intelligence business Transfer of learning Convolutional neural network Classifier (UML) Visualization
Zdroj:	Computer Vision – ECCV 2020 ISBN: 9783030585914 ECCV (23)
DOI:	10.1007/978-3-030-58592-1_2
Popis:	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.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::e194cdcf3e8b4612fe82f75cf3a13ce2 https://doi.org/10.1007/978-3-030-58592-1_2 Zobrazit plný text záznamu