It Contains two files.One is for data preparation(Preparation Dataset.ipynb) as the directory structure is Bit complex. And the other one is for traning and testing(Contains Deep Res Net)(3D_reconstructVNET.ipynb).
DataSet is taken from 3D R2N2 Research Paper.
They used ShapeNet models to generate rendered images and voxelized models. Links are available below.
ShapeNet rendered images http://cvgl.stanford.edu/data2/ShapeNetRendering.tgz
ShapeNet voxelized models http://cvgl.stanford.edu/data2/ShapeNetVox32.tgz
We have our 3D models in the form of .binox files. So for the conversion to binary voxel array we have Python module binvox-rw-py.
Attaches the link-- https://github.com/dimatura/binvox-rw-py
