search

cs-chan / Total-Text-Dataset

Total Text Dataset. It consists of 1555 images with more than 3 different text orientations: Horizontal, Multi-Oriented, and Curved, one of a kind.
BSD 3-Clause "New" or "Revised" License
747 stars 142 forks source link

Misunderstanding the scripts #39

Closed Didier0 closed 2 years ago

Didier0 commented 2 years ago

Hi, can you explain me some scripts?

What is difference between: T3 annotation tool scripts, Detection_Recognition_Annotation_script.m and the python scripts in Evaluation_Protocol?

Some script is used for generating file with groud truth of text on image (T3 ?), one is used for drawing polygon bounding box (Detection_Recognition_Annotation_script.m?) and the main script where magic is happening is Deteval.py or Pascal_VOC.py ?

chunchet-ng commented 2 years ago

HI there,

I think you are misunderstanding the purpose of these scripts. Files in the Annotation Tool folder are meant for the annotation platform we used to annotate TotalText, while the Evaluation Protocol folder has necessary scripts to evaluate the prediction results from a model.

If you want to draw polygons on the image, you need to write your own code to handle:

  1. Reads in TotalText ground-truth polygons and image
  2. Draw on the image using Pillow or OpenCV

Do note that we do not provide such code in this repo.

Didier0 commented 2 years ago

Ok, thank you for explanation!