search

tomasjakab / imm

Code for unsupervised learning of object landmarks as proposed in "Unsupervised Learning of Object Landmarks through Conditional Image Generation", Tomas Jakab*, Ankush Gupta*, Hakan Bilen, Andrea Vedaldi, Advances in Neural Information Processing Systems (NeurIPS) 2018.
149 stars 25 forks source link

questions about the data processing for Human 3.6M #12

Open justimyhxu opened 4 years ago

justimyhxu commented 4 years ago

I have some questions about the preprocessing about Human 3.6M.

  1. what the relax ratio for crop about the human instance.
  2. In the training procedure, all images need to forward in one epoch or you only use a subset of the training images?
  3. Could you provide the dataloader file for Human3.6m and pretrain model? Many thanks.