UIT-ViIC: A Dataset for the First Evaluation on Vietnamese Image Captioning

Image Captioning, the task of automatic generation of image captions, has\nattracted attentions from researchers in many fields of computer science, being\ncomputer vision, natural language processing and machine learning in recent\nyears. This paper contributes to research on Image Captioning task in terms of\nextending dataset to a different language - Vietnamese. So far, there is no\nexisted Image Captioning dataset for Vietnamese language, so this is the\nforemost fundamental step for developing Vietnamese Image Captioning. In this\nscope, we first build a dataset which contains manually written captions for\nimages from Microsoft COCO dataset relating to sports played with balls, we\ncalled this dataset UIT-ViIC. UIT-ViIC consists of 19,250 Vietnamese captions\nfor 3,850 images. Following that, we evaluate our dataset on deep neural\nnetwork models and do comparisons with English dataset and two Vietnamese\ndatasets built by different methods. UIT-ViIC is published on our lab website\nfor research purposes.\n