fredfung007 / snlt

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

about raw results

lawpdas opened this issue · comments

Hi,
I want to cite this work and make some comparisons. Could you provide the raw results?
I also notice that you have included TNL2k in your code.
Have you re-trained the model using this dataset?
Could you provide the new model and the evaluation result on this dataset?

I didn't train the model on TNL2K. I will upload the raw results once I get back from my internship. (around September)

BTW, the TNL2K have a lot of scene cut cases, and the current model is not performing very well on those cases.