10 datasets found

A vision-language pre-training dataset, Chinese CLIP, which consists of 100 million image-text pairs.

A vision-language pre-training dataset, BLIP2, which consists of 100 million image-text pairs.

The ALADIN dataset is a custom dataset created for the ALADIN paper.

The WebLI dataset used for training and evaluation of the CoBIT model.

The JFT-4B dataset used for training and evaluation of the CoBIT model.

The ALIGN dataset used for training and evaluation of the CoBIT model.

The dataset used for training and evaluation of the CoBIT model, which consists of image-text pairs from large-scale noisy web-crawled data and image annotation data.

Visual Question Answering (VQA) has achieved great success thanks to the fast development of deep neural networks (DNN). On the other hand, the data augmentation, as one of the...

MixGen: a joint data augmentation for vision-language representation learning to further improve data efficiency.

Large scale datasets [18, 17, 27, 6] boosted text conditional image generation quality. However, in some domains it could be difficult to make such datasets and usually it could...