We present Imagen, a text-to-image diffusion model with an unprecedented degree of photorealism and a deep level of language understanding. site 
, arxiv
An art Gallery …
Imagen builds on the power of large transformer language models in understanding text and hinges on the strength of diffusion models in high-fidelity image generation.
Our key discovery is that generic large language models (e.g. T5), pretrained on text-only corpora, are surprisingly effective at encoding text for image synthesis: increasing the size of the language model in Imagen boosts both sample fidelity and image-text alignment much more than increasing the size of the image diffusion model. Imagen achieves a new state-of-the-art FID score of 7.27 on the COCO dataset, without ever training on COCO, and human raters find Imagen samples to be on par with the COCO data itself in image-text alignment. To assess text-to-image models in greater depth, we introduce DrawBench, a comprehensive and challenging benchmark for text-to-image models. With DrawBench, we compare Imagen with recent methods including VQ-GAN+CLIP, Latent Diffusion Models, and DALL-E 2, and find that human raters prefer Imagen over other models in side-by-side comparisons, both in terms of sample quality and image-text alignment.
Some adventure game fans at Google wondered what would happen if you used the output from the classic text adventure game "Zork" as the input to Imagen, Google's text-to-image diffusion model. To support the results we wanted we made some additions to Zork itself, which was a fascinating journey through an arcane language and a lost but now re-engineered toolchain.