Conditional Image-Text Embedding Networks

european conference on computer vision, 2018.

Cited by: 42|Bibtex|Views108|DOI:https://doi.org/10.1007/978-3-030-01258-8_16
EI
Other Links: dblp.uni-trier.de|academic.microsoft.com|arxiv.org

Abstract:

This paper presents an approach for grounding phrases in images which jointly learns multiple text-conditioned embeddings in a single end-to-end model. In order to differentiate text phrases into semantically distinct subspaces, we propose a concept weight branch that automatically assigns phrases to embeddings, whereas prior works predef...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments