ercanburak / hypere2vid Goto Github PK

Hello, i'm super interested on this project, i wanted to know if, starting with the data in the csv format, is it possible to use this model?

Thank you for your outstanding work! I would like to know when to release the training code, thank you.

Hi! I really appreciate your fantastic work. When I run your code in datagen folder to generate rosbag data, I encountered two warnings. It seems that I didn't mkdir two folders. But I checked the code and config files, I didn't get what to mkdir. I wonder if these warnings necessary. How can I fix it or can I just ignore it? Thank you for your kindest reply.

Warnings are here:
W0410 15:00:56.242570 3319555 adaptive_sampling_benchmark_publisher.cpp:57] Empty benchmark folder string: will not write benchmark files
W0410 15:00:56.242695 3319555 synthetic_optic_flow_publisher.cpp:41] Empty output folder string: will not write synthetic optic flow files

Hello, when I reproduced your article Hyber-E2VID, I noticed that you used the TBPTT algorithm during training. I would like to know how this algorithm is implemented. If possible, could you share your code about this part?

Hi, I have a quesiton about the training data. I guess the data is generated with the same method as this work: https://timostoff.github.io/20ecnn. However, I found it difficult to generate high-quality event data using its method. May I ask your plan about release the related train data? Thanks very much!

Hi, First of all thanks for this great work.

I want to know, when you are planning to release code for this work?