search

garyptchoi / rectangular-conformal-map

Rectangular conformal map for simply-connected open surfaces
Apache License 2.0
13 stars 0 forks source link

Singular matrix resulting in non-square map #2

Open alex997-phy opened 2 weeks ago

alex997-phy commented 2 weeks ago

Hi,

I am mapping a 3d mesh into a 2D square. Although some meshes work good, some of them shows warning " Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 1.110223e-16.> In rectangular_conformal_map (line 132) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 161) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 162)".

It should be the problem that the matrix M (M = cotangent_laplacian(v,f)) is singular. And as a result, the 2D map is no longer square and the angle distortion is large. I do not know what to do to prevent this problem happening. I have tried to smooth the mesh but nothing changes. Would you mind helping me in solving this problem?

Thanks, Alex

muzire commented 2 weeks ago

谢谢来信，您的邮件已收到。祝您健康愉快！

garyptchoi commented 2 weeks ago

Hi,

I am mapping a 3d mesh into a 2D square. Although some meshes work good, some of them shows warning " Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 1.110223e-16.> In rectangular_conformal_map (line 132) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 161) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 162)".

It should be the problem that the matrix M (M = cotangent_laplacian(v,f)) is singular. And as a result, the 2D map is no longer square and the angle distortion is large. I do not know what to do to prevent this problem happening. I have tried to smooth the mesh but nothing changes. Would you mind helping me in solving this problem?

Thanks, Alex

Hi Alex, it will be good to check whether there are any degenerate faces/small topological holes/bad boundary vertices in the input 3D mesh, as they may cause the singular matrix issue. If you can send me the mesh file via email, I will be happy to look into it further.

alex997-phy commented 2 weeks ago

Hi Gray,

Thanks you for the quick reply. Please find the attached .ply file of the mesh.

Thanks, Alex

发件人: Gary P. T. Choi @.> 发送时间: 2024年9月17日 3:23 收件人: garyptchoi/rectangular-conformal-map @.> 抄送: alex997-phy @.>; Author @.> 主题: Re: [garyptchoi/rectangular-conformal-map] Singular matrix resulting in non-square map (Issue #2)

Hi,

I am mapping a 3d mesh into a 2D square. Although some meshes work good, some of them shows warning " Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 1.110223e-16.> In rectangular_conformal_map (line 132) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 161) Warning: Matrix is singular to working precision. > In rectangular_conformal_map (line 162)".

It should be the problem that the matrix M (M = cotangent_laplacian(v,f)) is singular. And as a result, the 2D map is no longer square and the angle distortion is large. I do not know what to do to prevent this problem happening. I have tried to smooth the mesh but nothing changes. Would you mind helping me in solving this problem?

Thanks, Alex

Hi Alex, it will be good to check whether there are any degenerate faces/small topological holes/bad boundary vertices in the input 3D mesh, as they may cause the singular matrix issue. If you can send me the mesh file via email, I will be happy to look into it further.

― Reply to this email directly, view it on GitHubhttps://github.com/garyptchoi/rectangular-conformal-map/issues/2#issuecomment-2353733943, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AOCYIQQJUONJJLG4KVHHPDTZW4VZTAVCNFSM6AAAAABOJ5FCVCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNJTG4ZTGOJUGM. You are receiving this because you authored the thread.Message ID: @.***>