FM-IQA Multilingual Question Answering Dataset

FM-IQA stands for Freestyle Multilingual Image Question Answering. It is a question-answering dataset that contains 150,000 images and 310,000 freestyle Chinese question-answer pairs and their English translations. This dataset can be used to train and evaluate mQA models. The model can be used to answer questions about the content of an image. The answer can be a sentence, a phrase, or a word.