Leveraging background augmentations to encourage semantic focus in self-supervised contrastive learning

Open in new window