Dataset Groups Activity Stream Caption MNIST Caption MNIST is a synthetic image-text pair dataset built by filling in the missing colors, digits, and positions in the MNIST dataset. BibTex: @dataset{Hyungyung_Lee_and_Sungjin_Park_and_Joonseok_Lee_and_Edward_Choi_2024, abstract = {Caption MNIST is a synthetic image-text pair dataset built by filling in the missing colors, digits, and positions in the MNIST dataset.}, author = {Hyungyung Lee and Sungjin Park and Joonseok Lee and Edward Choi}, doi = {10.57702/0p137mcb}, institution = {No Organization}, keyword = {'Image-Text Pair Generation', 'Multimodal Learning'}, month = {dec}, publisher = {TIB}, title = {Caption MNIST}, url = {https://service.tib.eu/ldmservice/dataset/caption-mnist}, year = {2024} }
