HyperAI

iVQA Video Question Answering Dataset

Download Help
特色图像

iVQA, short for Instructional Video Question Answering, is an open video question answering dataset with reduced language bias and high-quality manual annotations.

The dataset contains 10,000 video clips, each of which contains 1 question and 5 answers, and each question must be answered with the help of the video.