Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MeshAnything – Converts 3D representations into efficient 3D meshes

buaacyw.github.io

Post reply on HN
Print view View original

11–20 of 71 posts

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#11

Post by jgord »

Certainly a lot of scope for this kind of thing .. people who do lidar scans or photogrammetry of buildings tend to end up with very large meshes or very large point clouds.. which means they need souped up PCs and expensive software to wrangle into some usable CAD format.

Its an area where things can be improved a lot imho - I did some work a while back fitting flat planes to pointclouds, and ended up with mesh model anything from 40x to 100x smaller data than the ptcloud dataset. see quato.xyz for samples where you can compare the cloud, the mesh produced.. and view the 3D model in recent browsers.

My approach had some similarity to gaussian splats... but using only planar regions .. great for buildings made of flat slabs, less so for smooth curves and foliage.

Applying their MeshAnything algo to fine meshes from photogrammetry scans of buildings would be of great benefit - probably getting those meshes down to a size where they can be shared as 3D webgl/threejs pages.

Even deciding on triangle points to efficiently tesselate / cover a planar region with holes etc, is basically a knapsack problem, which heuristics, monte-carlo and ML can improve upon.

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#12

Post by spookie »

jgord wrote: post #11

Certainly a lot of scope for this kind of thing .. people who do lidar scans or photogrammetry of buildings tend to end up with very large meshes or very large point clouds.. which means they need souped up PCs and expensive software to wrangle into some usable CAD format. Its an area where things can be improved a lot imho - I did some work a while back fitting flat planes to pointclouds, and ended up with mesh mode…

Could you go more in depth with your approach? Sounds really usefull for man made structures!

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#13

Post by spookie »

wildpeaks wrote: post #10

Calling AI-generated meshes "Artist-created" just because it aims to look similar as human-made ones is misleading.

I think it comes from the fact an artist can create the mesh themselves and optimize them using this approach.

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#15

Post by spookie »

Joel_Mckay wrote: post #7
iTokio wrote: post #6

Words have meanings, you can’t call AI generated meshes, “Artist created Meshes” not matter how good you think your results are. Beside good topology is dependent on the use case, it’s very different if you are doing animation, a 3D print, a game or just a render.

Yep, hard to reason with industry people pushing slop on commercial production teams. Low-poly re-mesh tools have been around for ages (some better than others), but there are good reasons pro's still do this step manually all the time. Primarily "good" is based on _where_ the quads, loops, and unavoidable n-gons end up in the model (or stuff ends up looking retro 90's.) There is also the complex legal side of algori…

That's fair, as someone pretty proficient in 3D modelling I understand your point. However, it also boils down to the scale of the project.

Imagine recreating part of real life city, creating a digital twin, for scientific purposes (testing human behaviour in fire hazards, or simply iterating on better park planning and road design for greater perceived safety). There's a lot to be done, and it's difficult to use procedural building methods if your aim is for people to recognize that area.

I'm making such a thing myself, purely academic, but god I wish I could speed things up.

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#17

Post by jgord »

spookie wrote: post #12
jgord wrote: post #11

Certainly a lot of scope for this kind of thing .. people who do lidar scans or photogrammetry of buildings tend to end up with very large meshes or very large point clouds.. which means they need souped up PCs and expensive software to wrangle into some usable CAD format. Its an area where things can be improved a lot imho - I did some work a while back fitting flat planes to pointclouds, and ended up with mesh mode…

Could you go more in depth with your approach? Sounds really usefull for man made structures!

ahh.. looking for investment, so reticent to talk about the internals.

Basically does a lot of matmulls and sampling to find good good planar fits...

Most of these sample datasets took an hour to compute, using 12 threads on a midrange CPU... should be doable in seconds if ported to a highend GPU.

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#19

Post by librasteve »

jgord wrote: post #17
spookie wrote: post #12

Earlier quoted context omitted.

Could you go more in depth with your approach? Sounds really usefull for man made structures!

ahh.. looking for investment, so reticent to talk about the internals. Basically does a lot of matmulls and sampling to find good good planar fits... Most of these sample datasets took an hour to compute, using 12 threads on a midrange CPU... should be doable in seconds if ported to a highend GPU.

suggest a nanopivot, make it work on Apple AI (if you haven’t already) and position as iPhone AI mesh

Re: MeshAnything – Converts 3D representations into efficient 3D meshes

#20

Post by emilern »

column wrote: post #18
emilk wrote: post #14

You can run it yourself here: https://huggingface.co/spaces/rerun/InstantMesh

I didn't find a way to download the mesh once generated. Is that option not available at all?

Good question. You can if you use this instead: https://huggingface.co/spaces/TencentARC/InstantMesh
Post reply on HN