DeepMind Auto-Generates Images from Textual Descriptions
I’m still digesting this one but wanted to share it in the mean time. DeepMind is building algorithms that auto-generate imagery based on textual cues. Input “orange-bodied, black-headed bird and it generates a small, photo-realistic image of just such a bird.
More background here:
I’m trying to track down the original paper and will update this post should I locate it.