DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Models | IEEE Conference Publication | IEEE Xplore