no code implementations • 23 Dec 2020 • Guodong Zeng, Till D. Lerch, Florian Schmaranzer, Guoyan Zheng, Juergen Burger, Kate Gerber, Moritz Tannast, Klaus Siebenrock, Nicolas Gerber
In this paper, we propose intra- and cross-modality semantic consistency (ICMSC) for UDA and our key insight is that the segmentation of synthesised images in different styles should be consistent.