OpenAI DALL-E - Creating Images from Text

Here's some up to the minute information on the latest deep learning generative model architecture from OpenAI. It's a transformer architecture based off of GPT-3 that allows the generation of high quality images from a textural description.  

We will first turn to a very well put together and extremely timely video tutorial by Yannic Kilcher .  We also have links to the 2 OpenAI blog posts as well.  And of course some HTC observations.


So like Yannic says in the video tutorial, some of his commentary is somewhat speculative since there is not a paper to reference for the specific implementation details.  Although i think he did an excellent job of explaining what is presented in the blog posts.

Speaking of the blog posts, here they are.

'DALL-E: Creating Images from Text is here.

The reference section in this blog post has some good links to other generative 'text to image' arooraches to check out.

'CLIP: Connecting Text and Images' is here.

Paper and code links in the blog post.


Observations

1: The VQ-VAE model architecture is a part of our upcoming Generative Model Deep Dive post (it's a part ofJustin Johnson's second lecture features in part 2).

2:  As a texture synthesis model, this could be pretty slick.  Not that it isn't already.

3:  Obviously we need to get on posting our upcoming coverage of transformer models.  Including transformer models for image processing.  I would not have anticipated auto-regressive models doing so well (as contrasted with other generative approaches).

4:  I'm sure there is a way to frame this within Yann LeCun's grand unified theory of generative models.  Worth thinking about because i'm sure there are ways to tweak it to potentially speed it up and make it more flexible for artists.

5:  No code for DALL-E, there is code for CLIP.  Would be a shame if OpenAI just sells this to Microsoft like they did GPT-3.  Although i'm sure we'll see a cast of thousands post their implementations of it soon on GitHub, so we await that anxiously.

6:  I was not aware of that ImageNet Sketch dataset.  Interesting.


Comments

Post a Comment

Popular posts from this blog

CycleGAN: a GAN architecture for learning unpaired image to image transformations

Image
 Like the Doublemint Twins touting the joys of Doublemint gum, 2 GANs are surely better than 1 GAN.  Especially if we package them together inside of one meta GAN module.  And this is exactly what the CycleGAN architecture does. Have you ever harbored dark secrets of turning a horse into a zebra?  The CycleGAN was developed to do just that. Learn how to turn a horse into a zebra.  And more. Now right away you can notice a difference between the image to image transformation GAN architectures we've been discussing over the last few posts.  Those last few posts described systems that learn from a database of matched input-output image pairs. And if your goal is to turn an edge representation into a nicely filled in continuous tone image, it's easy to build your database of matched input-output image pairs that your GAN system can then learn off of.  Take a continuous tone image (which will be the output of the database pair entry), then run it through an edge detector algorithm. 
Read more

Pix2Pix: a GAN architecture for image to image transformation

Image
 I thought following up yesterday's TraVelGAN post with a Pix2Pix GAN post would be useful to compare what is going on in the 2 architectures.  Two different approaches to the same problem. I stole this Pix2Pix Overview slide below from an excellent deeplearning.ai GAN course (note that they borrowed it from the original paper) because it gives you a good feel for what is going on inside of the Pix2Pix architecture.    Note how the Generator part is very much like an auto-encoder architecture, but rebuilt using the U-Net architecture features (based on skip-connections) that fastai has been discussing in their courses for several years before it became more widely known to the deep learning community at large  (and which originally came from an obscure medical image segmentation paper) . So the Generator in this Pix2Pix GAN is really pretty sophisticated, consisting of a whole image to image auto-encoder network with U-Net skip connections to generate better image quality at highe
Read more

Smart Fabrics

Image
"What makes smart fabrics revolutionary is that they have the ability to do many things that traditional fabrics cannot, including communicate, transform, conduct energy and even grow."  This is a quote from Rebeccah Pailes-Friedman, who is a professor at Pratt with a research focus on developing wearable technology and smart textiles. Enhancing textiles can be viewed through different lens.  Enhancements can be aesthetic, or performance boosting, or durability boosting. 
Read more