Hierarchical text-conditional image generation with CLIP latents

4 years ago 35
Read Entire Article