Multi30K
Introduced 2016-05-02
Multi30K is a large-scale multilingual multimodal dataset for interdisciplinary machine learning research. It extends the Flickr30K dataset with German translations created by professional translators over a subset of the English descriptions, and descriptions crowdsourced independently of the original English descriptions. The dataset was introduced to stimulate multilingual multimodal research.
Benchmarks
Machine Translation/BLEU (EN-DE)Machine Translation/BLUE (DE-EN)Machine Translation/Meteor (EN-DE)Machine Translation/Meteor (EN-FR)Multimodal Machine Translation/BLEU (EN-DE)Multimodal Machine Translation/BLUE (DE-EN)Multimodal Machine Translation/Meteor (EN-DE)Multimodal Machine Translation/Meteor (EN-FR)