search

MichalBusta / E2E-MLT

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

Just curious what are the bbox values in the raw dataset labels? #46

Open zhenglilei opened 5 years ago

zhenglilei commented 5 years ago

The MLT format labels have been given. Thanks a lot!

But i am curious what do these floats in the raw label mean? 0 0.46809631347656255 0.8593862826680818 0.0739853187168 0.0808596013931 -0.0298556635598 منذ

The first number is the language label, e.g., 0 for Arabic, 2 for Chinese... The 2nd and 3rd number is the center point of the bounding box. But I have no idea about the others, thanks in advance if someone can clear my question.

MichalBusta commented 5 years ago

Lang,centerX,centerY,width,height,angle,transcripion the width and height are divided by image diagonal.

pá 19. 7. 2019 v 23:10 odesílatel zhenglilei notifications@github.com napsal:

The MLT format labels have been given. Thanks a lot!

But i am curious what do these floats in the raw label mean? 0 0.46809631347656255 0.8593862826680818 0.0739853187168 0.0808596013931 -0.0298556635598 منذ

The first number is the language label, e.g., 0 for Arabic, 2 for Chinese... The 2nd and 3rd number is the center point of the bounding box. But I have no idea about the others, thanks in advance if someone can clear my question.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MichalBusta/E2E-MLT/issues/46?email_source=notifications&email_token=AA7KHMFTUM4TSJGJASHW7VDQAJ63BA5CNFSM4IFNM572YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HANHAIQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AA7KHMFMBNINAMZ6324WVVDQAJ63BANCNFSM4IFNM57Q .

ThisIsIsaac commented 5 years ago

@MichalBusta

I'm still a little confused.

How do we convert centerX, centerY, width, height, and angle to actual boxes?

I have guesses for how to interpret centerX, centerY, width, height:

box_center_x_coord = center_x * image_width
box_center_y_coord = center_y * image_height
box_width = width * image_diagonal
box_height = height * image_diagonal

But I have no idea what angle means. Is it rotation in radian? Where is the rotation centered?

MichalBusta commented 5 years ago

yes, it is rotation in radians centered in the center of bounding box. complete code:
https://github.com/MichalBusta/E2E-MLT/blob/c147bfbf7fa66c97419d2d38333cbb0f33e91492/data_gen.py#L64