Internal error: Closed surface not correctly partitioned

[Silverlan]

When trying to call DirectX::UVAtlasCreate with a specific mesh, it consistently fails with a Closed surface not correctly partitioned error. I have uploaded an example project (including the data for the mesh in question) to replicate the issue here: test_uvatlas.zip The mesh is rather large, so UVAtlasComputeIMTFromTexture takes a few minutes to complete.

It works fine with other, simpler meshes.

Edit: I've tried it with another large mesh and got the same error. I have no idea what to look for, otherwise I'd try to narrow it down. Also, DirectX::UVAtlasComputeIMTFromTexture can take a few hours to complete, I'm not sure if that's normal.

[walbourn]

There are a number of these warnings before the failure:

Try to clean the non-manifold mesh, generated by partition

The DirectXMesh Clean and Validate methods are very useful for conditioning the mesh before you attempt to use UVAtlas.

https://github.com/microsoft/DirectXMesh/wiki/DirectXMesh#mesh-cleanup-and-validation

[chaohuang]

There are a number of these warnings before the failure:

Try to clean the non-manifold mesh, generated by partition

The DirectXMesh Clean and Validate methods are very useful for conditioning the mesh before you attempt to use UVAtlas.

https://github.com/microsoft/DirectXMesh/wiki/DirectXMesh#mesh-cleanup-and-validation

@walbourn Hi Chuck,

I checked the codes and found that the UVAtlasTool already called the DirectXMesh Clean and Validate methods before calling the UVAtlasCreate, but for some meshes, I still got the "closed surface not correctly partitioned" error.

Does this mean there could be some non-manifold meshes that couldn't be fixed by the DirectXMesh Clean and Validate methods, and UVAtlasCreate would fail on those meshes?

Thanks!

[walbourn]

Right. meshconvert only calls Clean if you use -c, but uvatlas always calls it.

Do you get any warnings from Validate (which is called before Clean) or just from UVAtlas?

[chaohuang]

Right. meshconvert only calls Clean if you use -c, but uvatlas always calls it.

Do you get any warnings from Validate (which is called before Clean) or just from UVAtlas?

Yes, I got warnings from Validate saying "bowties found" and "vertex dups". But I thought Clean would take care of those issues. I also got those warnings when processing others meshes, but UVAtlasTool worked fine on those meshes.

[walbourn]

Try hacking up your local copy of uvatlas to do the Validate again after the Clean. It seems like it's not fixing all the issues...

[chaohuang]

Try hacking up your local copy of uvatlas to do the Validate again after the Clean. It seems like it's not fixing all the issues...

I put another Validate after Clean, and the message returned from the 2nd Validate is empty. I also test the Validate method with all available VALIDATE_FLAGS, and the results are the same: the Validate before Clean reports "bowties found" warning, and the one after Clean doesn't report anything.

[walbourn]

OK, so the Clean worked but there's some degenerate case in your model that's tripping up UVAtlas...