Sight-seeing in the eyes of deep neural networks

Sight-seeing in the eyes of deep neural networks

Khademi, S. (TU Delft Pattern Recognition and Bioinformatics)
Shi, X. (TU Delft Pattern Recognition and Bioinformatics)
Mager, Tino (TU Delft History, Form & Aesthetics)
Siebes, R.M. (TU Delft Pattern Recognition and Bioinformatics; Vrije Universiteit Amsterdam)
Hein, C.M. (TU Delft History, Form & Aesthetics)
De Boer, Victor (Vrije Universiteit Amsterdam)
van Gemert, J.C. (TU Delft Pattern Recognition and Bioinformatics)

Hazeleger, W. (editor)

2018

We address the interpretability of convolutional neural networks (CNNs) for predicting a geo-location from an image. In a pilot experiment we classify images of Pittsburgh vs Tokyo and visualize the learned CNN filters. We found that varying the CNN architecture leads to variating in the visualized filters. This calls for further investigation of the effective parameters on the interpretability of CNNs.

Classification
Convolutional neural network (CNN)
Interpretability
Place recognition
Visualization

http://resolver.tudelft.nl/uuid:c7b3733a-9848-4e55-996f-978b49c98a60

https://doi.org/10.1109/eScience.2018.00125

IEEE

2019-06-01

978-153869156-4

Proceedings - IEEE 14th International Conference on eScience, e-Science 2018

14th IEEE International Conference on eScience, e-Science 2018, 2018-10-29 → 2018-11-01, Amsterdam, Netherlands

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

conference paper

© 2018 S. Khademi, X. Shi, Tino Mager, R.M. Siebes, C.M. Hein, Victor De Boer, J.C. van Gemert