A new model for producing 3D images from OpenAI

A new model for producing 3D images from OpenAI
A new model for producing 3D images from OpenAI
-
OpenAI has unveiled its latest model, Shap-E, which allows creating 3D objects via text commands, the same way the company’s model for creating 2D images, DALL-E, works.

Shap-e is a conditional generative artificial intelligence model that generates 3D objects, after being trained on a combination of 3D designs and a conditional diffusion model, the company said in a statement.

This free program is more difficult to install and setup than the popular platform “ChatGPT”, as the operation process requires the user’s programming experience after downloading the model from the GitHub platform for free and accessing it on the Microsoft Paint 3D application.

Avram Belch, editor-in-chief of Tom’s Hardware, said that the installation and commissioning process took 8 hours, considering that the company provides little guidance on the operation process.

-

He added that he was able to use the new model to convert different types of images into three-dimensional figures, such as animated GIFs, and he also used the feature that allows users to download a 2D image and convert it to 3D.

-

Belch pointed out that those who are trying to install the model and create 3D models should keep in mind that the model consumes a lot of system resources from the computer, which affects its performance with other tasks.

The new model is compatible with graphics processing units from “NVIDIA” only, and requires high-performance central processing units, so that the user can create three-dimensional objects in a matter of minutes.

-

-

PREV Google retracts its decision … and announces the “Pixel Tablet”
NEXT Scientists have revealed a more diverse human genome