3d model from real images

[TheShasa]

Hello, thanks for this great work!

Can you give me a hint how can we create the 3d model same as in lego extract_mesh.ipynb example using real life images?

[Sch1nken]

If I understand the readme correct, then you could try this:

Extracting geometry from a NeRF

Check out extract_mesh.ipynb for an example of running marching cubes to extract a triangle mesh from a trained NeRF network. You'll need the install the PyMCubes package for marching cubes plus the trimesh and pyrender packages if you want to render the mesh inside the notebook:

pip install trimesh pyrender PyMCubes

[TheShasa]

Thanks for your reply So first we need to train it on our 'real images' example?

[bmild]

Yes, you would have to train a network on your real images. Extracting a mesh in this case is a bit more complicated since it's harder to separate the object from the background than with the synthetic data that is isolated in space.

[TheShasa]

Great, thank you for clarification!