longhoang01/video-frame-qa-v5

该数据集是一个包含问题、选项、图片、支持性帧、视频路径和答案的训练集。具体来说，每个样本都包括一个唯一标识符、一个问题文本、一组选项、一张相关图片、一组支持性图片帧以及一个视频文件的路径和正确答案。

The dataset is a training set containing questions, options, images, supportive frames, video paths, and answers. Specifically, each sample includes a unique identifier, a question text, a set of options, a related image, a set of supportive image frames, a path to a video file, and the correct answer.