Wsabie: Scaling Up To Large Vocabulary Image Annotation
Venue
Proceedings of the International Joint Conference on Artificial Intelligence, IJCAI (2011)
Publication Year
2011
Authors
Jason Weston, Samy Bengio, Nicolas Usunier
BibTeX
Abstract
Image annotation datasets are becoming larger and larger, with tens of millions of
images and tens of thousands of possible annotations. We propose a strongly
performing method that scales to such datasets by simultaneously learning to
optimize precision at the top of the ranked list of annotations for a given image
and learning a low-dimensional joint embedding space for both images and
annotations. Our method, called Wsabie, both outperforms several baseline methods
and is faster and consumes less memory.
