hzxie / Pix2Vox

The official implementation of "Pix2Vox: Context-aware 3D Reconstruction from Single and Multi-view Images". (Xie et al., ICCV 2019)

Home Page:https://haozhexie.com/project/pix2vox

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Custom dataset training with ShapeNet-like structure.

sobikim opened this issue · comments

Hi.
I have some questions for training using custom dataset(ShapeNet-like structure)

  1. If not 24 images per object, would it be a problem for training?
  2. If the image size isn't 137 x 137, would it be a problem for training?
  3. If the object isn't centered in the image, will it be a problem for training?

Thank you!

  1. No
  2. No
  3. No

Thank you very much!