Bas Peters

Different from most deep learning settings, we focus on the situation where we have only one or a few large-scale examples. We also assume that only partial labels are available. Examples include video segmentation and segmentation of 3D geophysical...