I've been on the lookout for tools in this area since I suspect they're going to be a large part of art development in the future -- start with something rough out of a tool like this, and use it to save yourself a bunch of time early on + work on the details.
Will share this!
Would you mind explaining how your partner aquires clients?
Are there online marketplaces for this type of work?
It's definitely not Fiverr or whatever typical thing you'd expect, it's hard to earn money that way.
Easier if you do niche stuff and cater to specific online communities + establish a reputation.
I suspect half the use case of this is "this does a quick and dirty shading that gets you 75% of the way there" but man, I would paint over everything about those sword examples.
Artist and Designers have many ideas, but limited time and skills to realize their ideas. Our vision is to build ML tools that shorten the distance from having an idea and bringing it to life.
Currently, we have 2 products - generator and SKR. SKR automatically renders 2d sketches. Generator generates abstract images meant to spark creativity.
Any and all feedback is appreciated!
Focus on doing one thing well as a tool. Leave creativity for people.
> Leave creativity for people
That seems like a very black and white way of thinking. Where do you draw the line between human and 'machine-assisted' creativity? The second you start creating a mood board for your next project, you've probably already taken advantage of algorithmic feeds and content aggregators, but it's still you feeding the machine with querys etc.
Algorithmic feeds and content aggregators are for kids or the next cheap "decorator" with designers title. There is a only one historically proven way to be good in something and it is called learning and practicing.
We're developing hardware prototypes with our industrial designers, and I get 3D CAD designs from them. They can take the time to turn the CAD into visual design, but that takes time. Would this be a good tool for them to take the CAD, take a photo, and create demonstrations for how the product would look in the real world?
Curious if that's a realistic direction, and how else you see it being used. Nice work.
A natural evolution of this I feel like would be going from 2D sketches and synthetically reconstructing them in 3D which would accelerate the entire design process as a whole . Something we're working on now.
As of now we function in 2D ideation phase of art & design. I hope this answers some of your thoughts and questions . Thanks for the comment.
Have you by chance seen any of the work Facebook has been doing related to the 2d to 3d conversions?
https://github.com/facebookresearch/pifuhd
In an ideal world (for me) the output would be an .fbx file that could be uploaded directly into blender or unreal engine.
I wonder if, in 5 years, we'll start to see this kind of trippy AI aesthetic filter into illustration and animation. Or maybe the AI will just get more naturalistic.
https://open.starmony.io/play/9f5ce8d6-cfac-4bd4-83b7-985851...
