contrastive language-image pretraining (CLIP)
A model trained on paired images and captions so that visual content can be matched to text descriptions.
A model trained on paired images and captions so that visual content can be matched to text descriptions.