Trellis – 3D mesh generative model
21–30 of 84 posts
Re: Trellis – 3D mesh generative model
#22Wow, these look amazing. I'm a layman, but I think this is what everyone's been thinking about ever since the first NeRF demos? EDIT: I went looking for those threads and found my own comment wishing for this 5 years ago https://news.ycombinator.com/item?id=22642628 The next step is to automatically add "nodes" to the 3D images where the model can pivot, rotate and whatnot and then boom, you have on-demand animated,…
The next step is to generate models with higher quality mesh topology that allows animation and editing without breaking the mesh. I've done a lot of retopologizing and if I (or AI) were to rig these models as-is there would be all kinds of shading and deformation issues. Even without animating they are glaringly triangulated up close. But I suspect really high quality 3D asset generation is just around the corner because all you'd have to do is join up the approach seen here with AI quad re-meshing based on estimated direction fields and feature detection, which is also getting scarily good.
Re: Trellis – 3D mesh generative model
#23Looks like the generated wireframes are still terrible and would need to be redone by hand. So no need for artists to worry about their job security just yet.
From a performance pov I've imported ridiculous meshes not intended for games into unreal. It worked.
I can see this being ok for anything other than triple A games right?
Re: Trellis – 3D mesh generative model
#24I can see the potential, but the images I give it must be very far outside of its training because all it generates are weird flat planes
Yea another miracle tool... Until you test it.
The result was pretty good for the mesh, at least 100x faster than having to do it from scratch
Re: Trellis – 3D mesh generative model
#25Re: Trellis – 3D mesh generative model
#26Looks like the generated wireframes are still terrible and would need to be redone by hand. So no need for artists to worry about their job security just yet.
Question: How much do crappy wireframes actually matter? From a performance pov I've imported ridiculous meshes not intended for games into unreal. It worked. I can see this being ok for anything other than triple A games right?
Re: Trellis – 3D mesh generative model
#27I was hoping you could upload several images from different angles to help it, but that doesn't appear to be a feature.
Re: Trellis – 3D mesh generative model
#28Wow, these look amazing. I'm a layman, but I think this is what everyone's been thinking about ever since the first NeRF demos? EDIT: I went looking for those threads and found my own comment wishing for this 5 years ago https://news.ycombinator.com/item?id=22642628 The next step is to automatically add "nodes" to the 3D images where the model can pivot, rotate and whatnot and then boom, you have on-demand animated,…
>The next step is to automatically add "nodes" to the 3D images where the model can pivot, rotate and whatnot and then boom, you have on-demand animated, interactive content. The next step is to generate models with higher quality mesh topology that allows animation and editing without breaking the mesh. I've done a lot of retopologizing and if I (or AI) were to rig these models as-is there would be all kinds of shad…
Re: Trellis – 3D mesh generative model
#29Very cool. I wonder if the fact that it featurizes 3d objects using voxels can be memory intensive considering only the surface voxels are actually relevant for 3d structure makin'
Octrees are memory-efficient for this kind of thing.