1 paper · 1 filter
Matt Angus, Mohamed ElBalkini, Samin Khan +5
In training deep neural networks for semantic segmentation, the main limiting factor is the low amount of ground truth annotation data that is available in currently existing datas…