Today's post is a part of our code series, sharing with you the new repositories, codes and startups that are doing some cool work in the domain of AI.

Stable Dreamfusion

Last quarter has seen things that were just possible in dreams or imagination coming to real life. With transformers producing text, images and now recently videos now this new repository helps you create 3D objects from the text.

Ain't it cool? and scary at the same time?

The GitHub repo helps you do that.

GitHub - ashawkey/stable-dreamfusion: A working implementation of text-to-3D dreamfusion, powered by stable diffusion.
A working implementation of text-to-3D dreamfusion, powered by stable diffusion. - GitHub - ashawkey/stable-dreamfusion: A working implementation of text-to-3D dreamfusion, powered by stable diffus...
GitHubashawkey

It is a PyTorch implementation of the text-to-3D model Dreamfusion, powered by the Stable Diffusion text-to-2D model.

It is based on the following paper

DreamFusion: Text-to-3D using 2D Diffusion
Recent breakthroughs in text-to-image synthesis have been driven by diffusionmodels trained on billions of image-text pairs. Adapting this approach to 3Dsynthesis would require large-scale datasets of labeled 3D data and efficientarchitectures for denoising 3D data, neither of which currently exi…
arXiv.orgBen Poole

The following 3D image is generated using the text :

a high quality photo of a pineapple

0:00
/

Now you can generate some of your own items in 3D just by typing text.

Let me know what are you producing with this.

Do you like our work?
Consider becoming a paying subscriber to support us!

Subscribe Now