This dataset contains videos of participants of various ages; slightly tilting their faces left, right, up and down. Usually used to train face recognition models.
Dataset containing open-ended questions about images. These questions require an understanding of vision, language and commonsense knowledge to answer.