ABCNet

paper. The paper will be updated soon.

We provide demo in this repository. We will update our code at adet.

Run Demo

bash vis_demo.sh

We assume that your symlinked datasets/totaltext directory has the following structure:

totaltext
|_ test_images
|  |_ 0000000.jpg
|  |_ ...
|  |_0000299.jpg
|_ annotations
|  |_ total_test.json 

Model [Google Drive]

Totaltext test data [Google Drive]

Syntext-150k [link] (Part1: 54,327 images. Part2: 94,723 images.)

Description

Generated Results Download

Bezier-curve generated script link.

CTW1500 visualization results [link] (original training images can be downloaded Here)

Totaltext visualization results [link]

Cite

@article{liu2020abcnet,
  title   =  {ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network},
  author  =  {Liu, Yuliang* and Chen, Hao* and Shen, Chunhua and He, Tong and Jin, Lianwen and Wang, Liangwei},
  journal =  {arXiv preprint arXiv:2002.10200},
  year    =  {2020}
}

• Authors contributed equally.

Copyright

Any suggestion is welcome. Please send email to liu.yuliang@mail.scut.edu.cn or yuliang.liu@adelaide.edu.au

For commercial purpose usage, please contact Dr. Lianwen Jin: eelwjin@scut.edu.cn

Copyright 2019, Deep Learning and Vision Computing Lab, South China China University of Technology. http://www.dlvc-lab.net