Authors:
Clemens Seibold
1
;
Johannes Künzel
1
;
Anna Hilsmann
1
and
Peter Eisert
1
;
2
Affiliations:
1
Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute, HHI, Einsteinufer 37, 10587 Berlin, Germany
;
2
Visual Computing Group, Humboldt University Berlin, Unter den Linden 6, 10099 Berlin, Germany
Keyword(s):
Segmentation, Classification, LRP, Relevance, Annotation.
Abstract:
The new era of image segmentation leveraging the power of Deep Neural Nets (DNNs) comes with a price tag: to train a neural network for pixel-wise segmentation, a large amount of training samples has to be manually labeled on pixel-precision. In this work, we address this by following an indirect solution. We build upon the advances of the Explainable AI (XAI) community and extract a pixel-wise binary segmentation from the output of the Layer-wise Relevance Propagation (LRP) explaining the decision of a classification network. We show that we achieve similar results compared to an established U-Net segmentation architecture, while the generation of the training data is significantly simplified. The proposed method can be trained in a weakly supervised fashion, as the training samples must be only labeled on image-level, at the same time enabling the output of a segmentation mask. This makes it especially applicable to a wider range of real applications where tedious pixel-level label
ling is often not possible.
(More)