Corrigendum to "3D shape segmentation via shape fully convolutional networks" [Computers & Graphics 70 (2018) 128-139]