[yil8]

You could use more levels of convolution with larger receptive field. But this corresponds to larger patches, e.g. 512x512 pixels, and larger patches sometimes may not just be pure tumor cells or pure normal cells. And if you are just predicting 1 label for larger patches, it sometimes confuses the learning. What we propose with CRF, is larger receptive field with dense predictions, e.g. predicting more than one labels, and we use CRF to model the correlation between labels.