forceflow/trimesh2

C++ library and set of utilities for input, output, and basic manipulation of 3D triangle meshes

I'm Jeroen Baert. I write software, mostly for graphics-related applications. I am the developer of libmorton. I'm part of the Nerdland podcast team. I'm most proficient in C/C++, CUDA, Python and Java.

• Blog (EN)
• Homepage (NL)
• Twitter
• Mastodon

trimesh2's Issues

Mesh crop into a given box

Hello, I find the function of "MESH crop" is just removing these triangles and vertices outside the box, the boundary of the resulting mesh maybe not get straight along the box...

icp error

Crash occurs when median_dist is 0

icp crash

hi,
when I test icp, got a crash bug, my test data link is
https://drive.google.com/open?id=1mRPrH8sSt3ozoWh6O3a0ejFEh4PnBs7l

It would be great to distribute this on Winget/Chocolatey

This is a great tool! I'd love to be able to install/manage it through Winget instead of another MSI to keep track of and install of various machines. Would that be possible?

Can you explain some datails of the curvature

libsrc/TriMesh_curvature.cc, starting from line 194th:

The following Ax=b would be solved. I just want to know where the left matrix come from? Why is it three rows and three collonms? Thanks.

uu uv 0	x1	dnu*u
0 uu+vv u*v	x2	dnuv + dnvu
0 0 v*v	x3	dnv*v

error LNK2038 when trying to use this lib

Hello,
I'm having some linker issues in using this library. Any suggest on how to fix it?
Sorry if I am not in the right place to ask.

5>..\..\..\utilsrc\mesh_align.cc(16): fatal error C1083: Cannot open include file: 'unistd.h': No such file or directory

Hi,
Could you please help me ?

My OS is Windows 10 & i'm using VS 15.

thnx.

Variable corruption crash with ICP

Hi,

When upgrading from 2.14 to 2.15 we got a crash when using the ICP method.
I've build the libs with use of VS 2017.

The crash is located on line 374 of lineqn.h with the exception: Run-Time Check Failure #2 - Stack around the variable 'e' was corrupted.

We are using the default, unmodified version of the Trimesh2 from this Github.

Callstack:

How can I render 3D model with texture?

Hello forceflow,
Thanks for your great work.

It helps me a lot.

How can I render 3D model with texture uv?

Could you give me an example?

Thanks!

Python version or implementation

Is there any chance that I can use this great c++ library under python? Do you plan to implement or release python version?

About the setup

Hi Jeroen,

Thank you for the works.

I am pretty new to Trimesh2. Could you please tell me how to setup trimesh2 in Visual Studio 2015?
(e.g., if I have an empty Win32 console project, what should I do to use trimesh2 features?)

Looking forward to your reply.

Best regards,
Lin

Strange translation results.

Hi Forceflow,

Firstly thank you for the great fork!

I think i've bug in the ICP algorithm.

With certain objects, when they are located over eachother, the translation get messed up, and the resulting transformation matrix will cause the object jump to the other side of the screen.
The resulting rotation matrix is perfect, but it is the translation that is causing trouble.

Its not present in all cases, but in the teeth (which are attached) the phenomenon is present.

What do you think about this?

testset.zip

Can I get the corresponding face id of a tstrip?

I want to do some filtering on faces. Is there an api to get the corresponding face id of a tstrip? Thanks!

TriMesh::cornerangle uses incorrect formula for angle from known dot-product.

according to a formula for dot product:
a_vec dot b_vec = a_mod * b_mod * cos(angle_a_b)

TriMesh::cornerangle misses a_mod * b_mod and the code should be:

inline float cornerangle(int i, int j)
{
using namespace ::std;

if (unlikely(faces.empty())) need_faces();
const point &p0 = vertices[faces[i][j]];
const point &p1 = vertices[faces[i][NEXT_MOD3(j)]];
const point &p2 = vertices[faces[i][PREV_MOD3(j)]];

// new lines down from here
const point &v1 = p1 - p0;
const point &v2 = p2 - p0;
const float m = len(v1) * len(v2);
// return line is replaced
return acos(v1.dot(v2) / m);
}

GLCamera::setupGL function

Hi,

I have a question about the GLCamera::setupGL() function. Based on the source code, it returns a projection matrix using:

diag = sqrt(float(sqr(width) + sqr(height)));
top = (float) height/diag * 0.5f*field_of_view * neardist;
bottom = -top;
right = (float) width/diag * 0.5f*field_of_view * neardist;
left = -right;
glFrustum(left, right, bottom, top, neardist, fardist)`.

However, the above calculation of (left, right, bottom, top) seems a bit different to what I have learned below (needs a tangent function):

function perspectiveCameraUsingFrustum (fovy, aspect, near, far) {
  top = near * Math.tan(toRadians(fovy)/2);
  bottom = -top;
  right = top * aspect;
  left = -right;
  return createFrustum(left, right, bottom, top, near, far);
}

Can anyone explain if they are the same or their differences ? Thanks in advance.

Is "utils/mesh_align.exe" use the method presented in paper named: "A Symmetric Objective Function for ICP"?

I want to make sure that whether "utils/mesh_align.exe" use the method presented in paper named: "A Symmetric Objective Function for ICP"?

If not, how can I use the method in the paper?

Thanks for your splendid work! It's very helpful.