您当前的位置:首页 > IT编程 > 数据集
| C语言 | Java | VB | VC | python | Android | TensorFlow | C++ | oracle | 学术与代码 | cnn卷积神经网络 | gnn | 图像修复 | Keras | 数据集 | Neo4j | 自然语言处理 | 深度学习 | 医学CAD | 医学影像 | 超参数 | pointnet | pytorch |

mlt_english+chinese 数据集

51自学网 2021-11-11 16:39:33
  数据集

text-detection-ctpn

Scene text detection based on ctpn (connectionist text proposal network). It is implemented in tensorflow. The origin paper can be found here. Also, the origin repo in caffe can be found in here. For more detail about the paper and code, see this blog. If you got any questions, check the issue first, if the problem persists, open a new issue.


NOTICE: Thanks to banjin-xjy, banjin and I have reonstructed this repo. The old repo was written based on Faster-RCNN, and remains tons of useless code and dependencies, make it hard to understand and maintain. Hence we reonstruct this repo. The old code is saved in branch master


roadmap

  •  reonstruct the repo
  •  cython nms and bbox utils
  •  loss function as referred in paper
  •  oriented text connector
  •  BLSTM

setup

nms and bbox utils are written in cython, hence you have to build the library first.

cd utils/bbox
chmod +x make.sh
./make.sh

It will generate a nms.so and a bbox.so in current folder.


demo

  • follow setup to build the library
  • download the ckpt file from googl drive or baidu yun
  • put checkpoints_mlt/ in text-detection-ctpn/
  • put your images in data/demo, the results will be saved in data/res, and run demo in the root
python ./main/demo.py

training

prepare data

  • First, download the pre-trained model of VGG net and put it in data/vgg_16.ckpt. you can download it from tensorflow/models
  • Second, download the dataset we prepared from google drive or baidu yun. put the downloaded data in data/dataset/mlt, then start the training.
  • Also, you can prepare your own dataset according to the following steps.
  • Modify the DATA_FOLDER and OUTPUT in utils/prepare/split_label.py according to your dataset. And run split_label.py in the root
python ./utils/prepare/split_label.py
  • it will generate the prepared data in data/dataset/
  • The input file format demo of split_label.py can be found in gt_img_859.txt. And the output file of split_label.py is img_859.txt. A demo image of the prepared data is shown below.


train

Simplely run

python ./main/train.py
  • The model provided in checkpoints_mlt is trained on GTX1070 for 50k iters. It takes about 0.25s per iter. So it will takes about 3.5 hours to finished 50k iterations.

some results


返回列表
NLP数据库
51自学网,即我要自学网,自学EXCEL、自学PS、自学CAD、自学C语言、自学css3实例,是一个通过网络自主学习工作技能的自学平台,网友喜欢的软件自学网站。
京ICP备13026421号-1