Official PyTorch implementation of ODISE: Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models [CVPR 2023 Highlight]
Label and caption supervised pre-trained models for ODISE trained with the COCO training dataset.
Model license terms: Creative Commons — Attribution-NonCommercial-ShareAlike 4.0 International — CC BY-NC-SA 4.0.