Using Transformers to Do Image Captioning

Asicas also demonstrates that it's getting easier to do this kinds of research, just based on where we are conceptually, in terms of our techniques. Yet it's really interesting to see how this is a almost international battle across countries. There's also quite an active community of open source sort of hackers, woang to match with dowdy too, and ey're making a lot of progress on that front as well. So even if an open eye doesn't give public access to this level of quality for text to image, it's going to be available to just about everybody soon, it looks like.