Joint image and feature adaptative attention-aware networks for cross-modality semantic segmentation | Publicación