akanazawa / cmr

Project repo for Learning Category-Specific Mesh Reconstruction from Image Collections
MIT License
474 stars 84 forks source link

Simple Question about Data Preprocessing #13

Closed zeinaabuaisheh closed 4 years ago

zeinaabuaisheh commented 5 years ago

Hello, I would like to train a model on ShapeNet with does not have neither foreground masks nor semantic annotations, what should I do and in which format I should save the data after annotation?

akanazawa commented 4 years ago

I haven't worked with Shapenets so others can add, but I'm sure in ShapeNet you can get foreground masks since the ground truth 3D is available (any renderer would give you a mask), not sure about the semantic annotations. The focus of this work is on datasets without any ground truth 3D models, so we didn't really look into this.

Best,

A