Skip to content

ox-vgg/Text-Detect-Recognize

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 

Repository files navigation

Note: If you use this code, please cite:

Deng, Dan, Haifeng Liu, Xuelong Li, and Deng Cai. "Pixellink: Detecting scene text via instance segmentation." In Thirty-Second AAAI Conference on Artificial Intelligence. 2018.

Liu, Yang, Zhaowen Wang, Hailin Jin, and Ian Wassell. "Synthetically supervised feature learning for scene text recognition." In Proceedings of the European Conference on Computer Vision (ECCV), pp. 435-451. 2018.

-- Prerequistes:

--Download pretrained model:

Download : https://drive.google.com/open?id=1PuLCYVG457UOFzWHz4GuerTzWABZR0b6

DETECTION: unzip pixel_link_vgg_4s.zip into ${pixel_link_root}/model/

RECOGNITION: put 0_480000.pth into $attention_net_root$/

-- Usage:

DETECTION:

  1. Add the path of $pixel_link_root$/pylib/src to your PYTHONPATH

    export PYTHONPATH=${pixel_link_root}/pylib/src:$PYTHONPATH

  2. Test on Images in demo.txt:

./scripts/test_any.sh 0 model/conv3_3/model.ckpt-38055 $pixel_link_root$ /demo.txt

  1. It will generate detection result called Detection.pkl, which can be used as the input to the recognition model.

RECOGNITION:

  1. Recognize the text from the detected boxes:

python Recognition_yang.py --detection_path $pixel_link_root$/Detection.pkl --gpus 0

  1. The results will be saved in the file: $attention_net_root$/Recognition.pkl

About

Scene Text Detection and Recognition

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 64.8%
  • Jupyter Notebook 35.1%
  • Shell 0.1%