0

E2E-MLT

E2E-MLT -- 用于多语言场景文本的不受约束的端到端方法。「E2E-MLT - an Unconstrained End-to-End Method for Multi-Language Scene Text」

Github 資源庫
主要指標

Github星跟蹤圖

E2E-MLT

E2E-MLT -- 用于多语言场景文本的不受约束的端到端方法，代码库为：https://arxiv.org/abs/1801.09919

@@inproceedings{buvsta2018e2e,
  title={E2E-MLT-an unconstrained end-to-end method for multi-language scene text},
  author={Bu{\v{s}}ta, Michal and Patel, Yash and Matas, Jiri},
  booktitle={Asian Conference on Computer Vision},
  pages={127--143},
  year={2018},
  organization={Springer}
}

要求

预训练模型

e2e-mlt, e2e-mlt-rctw 

wget http://ptak.felk.cvut.cz/public_datasets/SyntText/e2e-mlt.h5

运行演示 

python3 demo.py -model=e2e-mlt.h5

数据

MLT SynthSet

合成文本是使用“Synthetic Data for Text Localisation in Natural Images（用于自然图像中文本本地化的合成数据）”生成的，对阿拉伯语和孟加拉语脚本渲染进行了较小的更改。

我们发现有用的东西：

训练 

python3 train.py -train_list=sample_train_data/MLT/trainMLT.txt -batch_size=8 -num_readers=5 -debug=0 -input_size=512 -ocr_batch_size=256 -ocr_feed_list=sample_train_data/MLT_CROPS/gt.txt

致谢

代码从 EAST 和 DeepTextSpotter 借用。


（The first version translated by vz on 2020.08.05）

主要指標

概覽
名稱與所有者MichalBusta/E2E-MLT
主編程語言C++
編程語言Python (語言數: 6)
平台Linux, Mac, Windows
許可證MIT License
所有者活动
創建於2018-10-16 13:12:19
推送於2025-01-08 09:27:36
最后一次提交2022-08-20 10:15:29
發布數0
用户参与
星數292
關注者數15
派生數84
提交數39
已啟用問題?
問題數76
打開的問題數31
拉請求數3
打開的拉請求數0
關閉的拉請求數0
项目设置
已啟用Wiki?
已存檔?
是復刻?
已鎖定?
是鏡像?
是私有?

E2E-MLT

E2E-MLT - an Unconstrained End-to-End Method for Multi-Language Scene Text
code base for: https://arxiv.org/abs/1801.09919

@@inproceedings{buvsta2018e2e,
  title={E2E-MLT-an unconstrained end-to-end method for multi-language scene text},
  author={Bu{\v{s}}ta, Michal and Patel, Yash and Matas, Jiri},
  booktitle={Asian Conference on Computer Vision},
  pages={127--143},
  year={2018},
  organization={Springer}
}

Requirements

Pretrained Models

e2e-mlt, e2e-mlt-rctw

wget http://ptak.felk.cvut.cz/public_datasets/SyntText/e2e-mlt.h5

Running Demo

python3 demo.py -model=e2e-mlt.h5

Data

MLT SynthSet

Synthetic text has been generated using Synthetic Data for Text Localisation in Natural Images, with minor changes for Arabic and Bangla script rendering.

What we have found useful:

  • for generating Arabic Scene Text: https://github.com/mpcabd/python-arabic-reshaper
  • for generating Bangla Scene Text: PyQt4
  • having somebody who can read non-latin scripts: we would like to thank Ali Anas for reviewing generated Arabic scene text.

Training

python3 train.py -train_list=sample_train_data/MLT/trainMLT.txt -batch_size=8 -num_readers=5 -debug=0 -input_size=512 -ocr_batch_size=256 -ocr_feed_list=sample_train_data/MLT_CROPS/gt.txt

Acknowledgments

Code borrows from EAST and DeepTextSpotter