search

MichalBusta / E2E-MLT

E2E-MLT - an Unconstrained End-to-End Method for Multi-Language Scene Text
MIT License
291 stars 84 forks source link

score_map vs. training_masks #43

Closed kei6 closed 5 years ago

kei6 commented 5 years ago

Dear @MichalBusta, I'm trying to understand the difference between score_map and training_masks, which are generated from: https://github.com/MichalBusta/E2E-MLT/blob/28583581fb17b6e83bc8dc8c84b6bc7fb4957341/data_gen.py#L555 Having some questions come in with me:

  1. Which one is more important to train the model? (I checked and found that score_map always have all ground-truth boxes, but training_masks is not)
  2. Why do you generate score_map and training_masks? (Because according to my work to understand your works, I think just score_map is enough. Pls help to fix if I have misunderstood, thanks!)
  3. Why do you just add a box to training_masks if its text contains " ", according to this line of code: https://github.com/MichalBusta/E2E-MLT/blob/28583581fb17b6e83bc8dc8c84b6bc7fb4957341/data_gen.py#L478-L496 Hope to hear from you soon, Thanks in advance,
MichalBusta commented 5 years ago

hi Kei, train mask Is used to mask out regions from score map - mostly:

we do not learn on those masked out parts.

úterý 2. července 2019 Kei_Cin notifications@github.com napsal(a):

Dear @MichalBusta https://github.com/MichalBusta, I'm trying to understand the difference between score_map and training_masks, which are generated from: https://github.com/ MichalBusta/E2E-MLT/blob/28583581fb17b6e83bc8dc8c84b6bc 7fb4957341/data_gen.py#L555 Having some questions come in with me:

  1. Which one is more important to train the model? (I checked and found that score_map always have all ground-truth boxes, but training_masks is not)
  2. Why do you generate score_map and training_masks? (Because according to my work to understand your works, I think just score_map is enough. Pls help to fix if I have misunderstood, thanks!)
  3. Why do you just add a box to training_masks if its text contains " ", according to this line of code: https://github.com/ MichalBusta/E2E-MLT/blob/28583581fb17b6e83bc8dc8c84b6bc 7fb4957341/data_gen.py#L478-L496 https://github.com/MichalBusta/E2E-MLT/blob/28583581fb17b6e83bc8dc8c84b6bc7fb4957341/data_gen.py#L478-L496 Hope to hear from you soon, Thanks in advance,

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/MichalBusta/E2E-MLT/issues/43?email_source=notifications&email_token=AA7KHMEJBOHH3YGS36X5Q4LP5OAHBA5CNFSM4H453452YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4G453FZA, or mute the thread https://github.com/notifications/unsubscribe-auth/AA7KHMGDUYUPPDCSEXQJDWDP5OAHBANCNFSM4H45345Q .

kei6 commented 5 years ago

Thanks for this quick reply! To be clear, the model learns from score_map, doesn't it?

MichalBusta commented 5 years ago

Yes, but only on parts given by train mask.

středa 3. července 2019 Kei_Cin notifications@github.com napsal(a):

Thanks for this quick reply! To be clear, the model learns from score_map, doesn't it?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/MichalBusta/E2E-MLT/issues/43?email_source=notifications&email_token=AA7KHMCOEEGZKAOO3LLTIPTP5RN6VA5CNFSM4H453452YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZDVFRI#issuecomment-507990725, or mute the thread https://github.com/notifications/unsubscribe-auth/AA7KHMGKWYM2SZ52V6F5ZYLP5RN6VANCNFSM4H45345Q .