-
Notifications
You must be signed in to change notification settings - Fork 66
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Procedure for voxel generation #19
Comments
Hi, Thank you for asking! We previously used some MATLAB code to generate voxels from mesh grids. Unfortunately, it is difficult for me to access the code after moving to Princeton. However, you may still use tools like binvox to voxelize meshes. Thank you! Best, |
Hi @xingyuansun
|
Hi, @purugupta99 I am going through the same procedure and am wondering if you figured it out? My obj file had xyz range of [-1, 1] but the coordinates stored in the binvox files have range approximately in [0, 256]. Is there any scaling we need to do before feeding the obj to binvox, or is there a way to get the R, T matrices? Edit: Nevermind, I figured it out. |
Hi, @ZZWENG , I'm facing the same issue, could you please kindly teach me how you figure it out? |
Hi, @purugupta99 , could you please tell me what program you are using to visualize the binvox files as in the upper picture? Edit: It is viewvox, I ignore that, nervermind. |
I am not sure what your use case is, so it may not be applicable to you. For me, I first convert all
And then, I used
|
Hi, @ZZWENG Thanks a lot for your kind reply! I'm trying to train meshrcnn on a custom 3D model dataset too. Could you please explain why you did these two operations when converting
|
I remember doing that while I was trying to replicate a pix3D voxel file using its .obj file. Maybe it is not necessary for you. |
I find it is a rotation of the coordinate system which was done in Pix3D, thank you very much!
… 2020年10月16日 上午12:36,Jen Weng ***@***.***> 写道:
Hi, @ZZWENG <https://github.com/ZZWENG> Thanks a lot for your kind reply! I'm trying to train meshrcnn on a custom 3D model dataset too. Could you please explain why you did these two operations when converting .binvox to .mat? Is them essential and what do they mean?
voxels = model.data.astype(np.int).transpose(0,2,1)
voxels[:, 1, :] = -voxels[:, 1, :]
I remember doing that while I was trying to replicate a pix3D voxel file using its .obj file. Maybe it is not necessary for you.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE5N4ZIFAWMJNVSUW7XKP3LSK4QJTANCNFSM4PKPW2CA>.
|
Hi, I know how to get the voxel.mat then, but i have a question about how to get the rot_mat and trans_mat ? |
Does anybody know the way to convert .mat to .binvox? |
Can you share the procedure for generating the voxels(.mat file) from the mesh files? I'm trying to generate similar dataset to be used for training MeshRCNN.
TIA
The text was updated successfully, but these errors were encountered: