On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation
Fig 6
Pixel-wise decomposition for Bag of Words features over χ2-kernels using the Taylor-type decomposition for the third layer and the layer-wise relevance propagation for the subsequent layers.
Left: The original image. Middle: Pixel-wise prediction. Right: Superposition of the original image and the pixel-wise prediction. The decompositions were computed on tiles of size 102 × 102 and having a regular offset of 34 pixels. The decompositions from the overlapping tiles were averaged. In the heatmap, based on linearly mapping the interval [−1, +1] to the jet color map available in many visualization packages, green corresponds to scores close to zero, yellow and red to positive scores and blue color to negative scores. See text for interpretation.