Computer Vision Values Dataset

This is a corpus of about 500 computer vision datasets, from which we sampled 114 dataset publications across different vision tasks and coded for themes through both structured and qualitative content analysis. This work most closely pairs with research question 1 in the genealogies of data project (https://arxiv.org/abs/2007.07399): How do dataset developers in CV and NLP research, describe and motivate the decisions that go into their creation?