google-research-datasets/wit

WIT (Wikipedia-based Image Text) Dataset is a large multimodal multilingual dataset comprising 37M+ image-text sets with 11M+ unique images across 100+ languages.

84 best points · 8 comments · 1 Hacker News post · 1,115 GitHub stars · View on GitHub

Hacker News discussions

• Wit: Wikipedia-Based Image Text Dataset (84 points, 8 comments, by jonbaer, 2021-03-04)