about network

[zhuyazhi]

hello~ I notice that in the paper there are three network to predict 3 kinds of intermidiate representations. but in the code, I only see there is one network which is the same as PVNet's, so I want to know if my understanding is true: use the same network in PVNet to predict two more intermidiate representations: egde vetors and symmetry correspondences? and add two correspoding loss to these two task? Another question is where can I change the training-test split in the code? wish your reply~thanks

[chensong1995]

Hi zhuyazhi,

Thanks for your interest in our work! We use three networks to predict 3 types of intermediate representations. However, most layers in these three networks are shared. Only the last layer is different.

We have updated the training-testing split today. Please check this out: https://github.com/chensong1995/HybridPose/blob/master/src/train_core.py#L50

[zhuyazhi]

Dear author, I noticed that you said the arXiv paper is updated, I went to load the paper and found it was the same as the prior version, could you tell me where can I get the updated paper? thasnk you~

[chensong1995]

Hi zhuyazhi,

I updated our arXiv manuscript on Thursday. The system notified me that "Your replacement is scheduled to be announced at Mon, 19 Oct 2020 00:00:00 GMT." The updated paper will be available in about 9 hours.

I hope this helps! Chen