The semantic segmentation task in computer vision involves partitioning an image into a set of multiple non-overlapping and semantically interpretable regions. This entails assigning pixel-wise class ...

This paper proposes an end-to-end trained fully convolutional neural network model to process 3D image volumes. Unlike previous works that processed the input volumes slice-wise or patch-wise, the ...