3 ms·
Super cool results. This is what my startup is getting into. So I'm very interested. These aren't "game ready" - the sculpts are pretty gross. But we're clear
by echelon 4y ago
Super cool results.
This is what my startup is getting into. So I'm very interested.
These aren't "game ready" - the sculpts are pretty gross. But we're clearly getting somewhere. It's only going to keep getting better.
I expect we'll be building all new kinds of game engines, render pipelines, and 3D animation tools shortly.
- nico 4y agoAnd 3D printing. So quickly building physical tools too.
- skybrian 4y agoFor printing parts, precision matters since they likely need to fit with something else. You’ll want to be able to edit dimensions on the model to get the fit right. So maybe someday, but I think it would have to be a project that targets CAD.
- bredren 4y agoHow do these kinds of tools complement actual 3d scanning? For example, Apple supposedly has put some time into 3d asset building (presumably in support of AR world building content). Can these inference techniques stack or otherwise help more detailed object data collection?
- redox99 4y agoWhile this is cool, this is not meant to target "game ready". For games and CGI, there's no reason to limit yourself to a single image. Photogrammetry is already extensively used, and it involves using tens or hundreds of images of the object to scan. Using many images as an input will obviously always be superior to a single one, as a single image means it has to literally make up the back side, and it has no parallax information.
- oefrha 4y agoYou appear to be thinking about scanning a physical object, whereas zero-shot one image to 3D object would be vastly more useful with a single (possibly AI-generated or AI-assisted) illustration. You get a 3D model in seconds at essentially zero cost, can iterate hundreds of times in a single day.
- redox99 4y agoI agree that for stylized, painting-like 3D models it could be very cool. I was indeed thinking of the typical pipeline for a photoreallistic asset.
- digilypse 4y agoWhat if I have a dynamically generated character description in my game’s world, generate a portrait for them using StableDiffusion and then turn that into a 3d model that can be posed and re-used?
- flangola7 4y agoThis has DARPA and NSF behind it. They're not building this for games they're building it for autonomous weapons.
- regularfry 4y agoI'd be interested in zero-shot two images to 3d object. You can see how a stereo pair ought to improve the amount of information it has available.