WebVidVQA3M

TextsVideosApache 2.0Introduced 2022-05-10

A dataset automatically generated using question generation neural models and alt-text video captions from the WebVid dataset, with 3M video-question-answer triplets.