Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Controlling explanatory heatmap resolution and semantics via decomposition depth

: Bach, S.; Binder, A.; Müller, K.-R.; Samek, W.


Karam, L. ; Institute of Electrical and Electronics Engineers -IEEE-; IEEE Signal Processing Society:
IEEE International Conference on Image Processing, ICIP 2016. Proceedings : September 25-28, 2016, Phoenix Convention Center, Phoenix, Arizona, USA
Piscataway, NJ: IEEE, 2016
ISBN: 978-1-4673-9962-3 (Print)
ISBN: 978-1-4673-9961-6 (Online)
International Conference on Image Processing (ICIP) <23, 2016, Phoenix/Ariz.>
Fraunhofer HHI ()

We present an application of the Layer-wise Relevance Propagation (LRP) algorithm to state of the art deep convolutional neural networks and Fisher Vector classifiers to compare the image perception and prediction strategies of both classifiers with the use of visualized heatmaps. Layer-wise Relevance Propagation (LRP) is a method to compute scores for individual components of an input image, denoting their contribution to the prediction of the classifier for one particular test point. We demonstrate the impact of different choices of decomposition cut-off points during the LRP-process, controlling the resolution and semantics of the heatmap on test images from the PASCAL VOC 2007 test data set.