Closed mmmmmore closed 4 years ago
Thanks!
It was intended to mark some potentially difficult text, such as text that are too small or text regions that are too dark.
Later I found that it will draw down the generation speed (especially when we need to check if a text region is too dark). Besides, in my initial experiments with the EAST models, filtering out these difficult samples did not help. So I took off the code, and just leave this flag there.
Thanks for open sourcing data generation code. May I know your the intended usage on the flag
is_difficult
?In the source code it is always set to be 0, this looks a bit confusing. Any additional explanation would be greatly appreciated !