Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models | IEEE Conference Publication | IEEE Xplore