Generative AI is the new thing right now, proving to be a useful tool both for professional programmers, writers of high school essays and all kinds of other applications in between. It’s also been shown to be effective in generating images, as the DALL-E program has demonstrated with its impressive image-creating abilities. It should surprise no one as this type of AI continues to make in-roads into other areas, this time with a program from OpenAI called Shap-E which can render 3D images.
Like most of OpenAI’s offerings, this takes plain language as its input and can generate relatively simple 3D models with this text. The examples given by OpenAI include some bizarre models using text prompts such as a chair shaped like an avocado or an airplane that looks like a banana. It can generate textured meshes and neural radiance fields, both of which have various advantages when it comes to available computing power, training methods, and other considerations. The 3D models that it is able to generate have a Super Nintendo-style feel to them but we can only expect this technology to grow exponentially like other AI has been doing lately.
For those wondering about the name, it’s apparently a play on the 2D rendering program DALL-E which is itself a combination of the names of the famous robot WALL-E and the famous artist Salvador Dali. The Shap-E program is available for anyone to use from this GitHub page. Even though this code comes from OpenAI themselves, plenty are speculating that the AI revolution to come will largely come from open-source sources rather than OpenAI or Google, something for which the future is somewhat hazy.
8 thoughts on “3D Design With Text-Based AI”
“chair shaped like an avocado” was a thing even before teh interwebs and AI, so generating tweens between that and actual avocadoes is facil. Try getting an AI to produce a decent image of an octopus wearing a space suit while repairing a satellite in orbit. You can see that in your mind right now can’t you, but an AI will always get it wrong in some significant way, particularly given that they all seem to be innumerate.
“I knew about this before it was cool.”
Not quite the same, but check https://www.kaedim3d.com/ out, generates full 3D meshes in very good quality from a 2D image using AI. Not open sourced in the slightest, and very expensive, but utterly incredible results, more so if combined with the DALL-E to create the source image.
Demo video – https://www.youtube.com/watch?v=1xsI-we37dM
I used them for a project, and quickly found out that they put up the images on a modeling-for-hire bid system and a real person does the work. Their billing and token system are whacked out too.
You’ll definitely get a model of the image(s) you sent in, but it’s almost certainly bidded out to humans.
Omfg, damn, you admin, can you mod this comment away for us all plz, definitely don’t want to be giving snakes like that any revenue. Dumbass me fell for the pretty marketing!
Leave as is, modding the original away would either delete the whole thread or make Db’s reply useless. Right now this serves as a warning to the community.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)