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 140 forks source link

请问这两种方式有何不同，是否得到相差较多的检测结果？ #40

Closed huahuabai closed 2 years ago

huahuabai commented 2 years ago

请问这两种方式有何不同，是否得到相差较多的检测结果？ 盼复，感谢您！

image

chunchet-ng commented 2 years ago

Hi there,

The differences between these two versions are as follows:

  1. Mask counting version is using skimage's polygon class, while another version is using shapely's.
  2. The later version has a faster speed, which was introduced by this pull request.

Theoretically, they should return the same results. We recommend you use the later version as it is more up-to-date.

Thank you.