search

ranahanocka / MeshCNN

Convolutional Neural Network for 3D meshes in PyTorch
MIT License
1.56k stars 314 forks source link

Problem in understanding the dataset #35

Closed ankitpatnala closed 4 years ago

ankitpatnala commented 4 years ago

i was working with your cube dataset to benchmark my dataset later. Each and every mesh had the same number of triangles and vertices which somehow won't be the case for me. Will it be possible to work on training data with varied number of triangles and vertices ?

ranahanocka commented 4 years ago

Hi @ankitpatnala ,

Are you working on classification or segmentation? The code supports varied number of faces (you must specify a number of max edges). However, I recommend simplifying the meshes to a target number of faces using the scripts I posted. Please read more in the data processing wiki and if you are doing segmentation so also check out the segmentation wiki

ankitpatnala commented 4 years ago

I am working on segmentation task. I want to know how can we create .seg and .eseg files as i am facing difficulty in understanding edge ordering . I got a particular arrangement of edges from the function fill_from_file in mesh_prepare.py. Do i need to maintain the same order ?

ankitpatnala commented 4 years ago

Do you have any algorithm that is suitable for instance segmentation on meshes ?

ranahanocka commented 4 years ago

Yes, you should maintain the edge order in fill_from_file.

About segmentation algorithms, see this issue

ranahanocka commented 4 years ago

closing due to lack of activity