search

kdexd / virtex

[CVPR 2021] VirTex: Learning Visual Representations from Textual Annotations
http://kdexd.xyz/virtex
MIT License
557 stars 61 forks source link

this model can be used in Multimodal retrieval problem #27

Closed yier2333 closed 2 years ago

kdexd commented 2 years ago

Yes, our model can be used for text-to-image retrieval. See this excellent paper by @antoine77340 et al, for more details — https://arxiv.org/abs/2103.16553

(This codebase does not have out-of-the-box support for retrieval.)