search

yuleiniu / vc

Code for CVPR'18 "Grounding Referring Expressions in Images by Variational Context"
29 stars 4 forks source link

Question about the input of regularizer #2

Open Sy-Zhang opened 5 years ago

Sy-Zhang commented 5 years ago

Hi Yulei,

Thanks for your great work. When reading your codes, I'm confused with the following lines: https://github.com/yuleiniu/vc/blob/86e7712879964fd5d03f9412e6ed0b442bc3bd26/vc_model.py#L295-L296 According to your paper, the input of regularizer is z but not x. Shouldn't these two lines look like these? 

region_visual_feat   = self.enc_visual_feat
region_spatial_feat  = self.enc_spatial_feat