Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models
This addresses the problem of segmenting any category in the wild for computer vision applications, representing a significant advance rather than an incremental improvement.
The paper tackles open-vocabulary panoptic segmentation by unifying pre-trained text-to-image diffusion and discriminative models, achieving state-of-the-art results with 23.4 PQ and 30.0 mIoU on ADE20K, an 8.3 PQ and 7.9 mIoU improvement over prior work.
We present ODISE: Open-vocabulary DIffusion-based panoptic SEgmentation, which unifies pre-trained text-image diffusion and discriminative models to perform open-vocabulary panoptic segmentation. Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse open-vocabulary language descriptions. This demonstrates that their internal representation space is highly correlated with open concepts in the real world. Text-image discriminative models like CLIP, on the other hand, are good at classifying images into open-vocabulary labels. We leverage the frozen internal representations of both these models to perform panoptic segmentation of any category in the wild. Our approach outperforms the previous state of the art by significant margins on both open-vocabulary panoptic and semantic segmentation tasks. In particular, with COCO training only, our method achieves 23.4 PQ and 30.0 mIoU on the ADE20K dataset, with 8.3 PQ and 7.9 mIoU absolute improvement over the previous state of the art. We open-source our code and models at https://github.com/NVlabs/ODISE .