liuheng92 / Tensorflow_psenet
Licence: mit
This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.My blog:
Stars: ✭ 472
Projects that are alternatives of or similar to Tensorflow psenet
East
A tensorflow implementation of EAST text detector
Stars: ✭ 2,804 (+494.07%)
Mutual labels: text-detection, ocr
vietnamese-ocr-toolbox
A toolbox for Vietnamese Optical Character Recognition.
Stars: ✭ 26 (-94.49%)
Mutual labels: ocr, text-detection
pytorch.ctpn
pytorch, ctpn ,text detection ,ocr,文本检测
Stars: ✭ 123 (-73.94%)
Mutual labels: ocr, text-detection
Dbnet.pytorch
A pytorch re-implementation of Real-time Scene Text Detection with Differentiable Binarization
Stars: ✭ 435 (-7.84%)
Mutual labels: text-detection, ocr
Chineseaddress ocr
Photographing Chinese-Address OCR implemented using CTPN+CTC+Address Correction. 拍照文档中文地址文字识别。
Stars: ✭ 309 (-34.53%)
Mutual labels: text-detection, ocr
Awesome Deep Text Detection Recognition
A curated list of resources for text detection/recognition (optical character recognition ) with deep learning methods.
Stars: ✭ 2,282 (+383.47%)
Mutual labels: text-detection, ocr
React Native Tesseract Ocr
Tesseract OCR wrapper for React Native
Stars: ✭ 384 (-18.64%)
Mutual labels: text-detection, ocr
East icpr
Forked from argman/EAST for the ICPR MTWI 2018 CHALLENGE
Stars: ✭ 154 (-67.37%)
Mutual labels: text-detection, ocr
Text Detection Ctpn
text detection mainly based on ctpn model in tensorflow, id card detect, connectionist text proposal network
Stars: ✭ 3,242 (+586.86%)
Mutual labels: text-detection, ocr
PSENet-Tensorflow
TensorFlow implementation of PSENet text detector (Shape Robust Text Detection with Progressive Scale Expansion Networkt)
Stars: ✭ 51 (-89.19%)
Mutual labels: ocr, text-detection
Text Detection
Text detection with mainly MSER and SWT
Stars: ✭ 167 (-64.62%)
Mutual labels: text-detection, ocr
Psenet.pytorch
A pytorch re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network
Stars: ✭ 416 (-11.86%)
Mutual labels: text-detection, ocr
Adelaidet
AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.
Stars: ✭ 2,565 (+443.43%)
Mutual labels: text-detection, ocr
Ocr.pytorch
A pure pytorch implemented ocr project including text detection and recognition
Stars: ✭ 196 (-58.47%)
Mutual labels: text-detection, ocr
Craft Pytorch
Official implementation of Character Region Awareness for Text Detection (CRAFT)
Stars: ✭ 2,220 (+370.34%)
Mutual labels: text-detection, ocr
doctr
docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning.
Stars: ✭ 1,409 (+198.52%)
Mutual labels: ocr, text-detection
Craft keras
Keras implementation of Character Region Awareness for Text Detection (CRAFT)
Stars: ✭ 143 (-69.7%)
Mutual labels: text-detection, ocr
Tedeval
TedEval: A Fair Evaluation Metric for Scene Text Detectors
Stars: ✭ 143 (-69.7%)
Mutual labels: text-detection, ocr
craft-text-detector
Packaged, Pytorch-based, easy to use, cross-platform version of the CRAFT text detector
Stars: ✭ 151 (-68.01%)
Mutual labels: ocr, text-detection
Megreader
A research project for text detection and recognition using PyTorch 1.2.
Stars: ✭ 332 (-29.66%)
Mutual labels: text-detection, ocr
PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network
Introduction
This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.
Thanks for the author's (@whai362) awesome work!
Installation
- Any version of tensorflow version > 1.0 should be ok.
- python 2 or 3 will be ok.
Download
trained on ICDAR 2015 (training set) + ICDAR2017 MLT (training set):
baiduyun extract code: pffd
This model is not as good as article's, it's just a reference. You can finetune on it or you can do a lot of optimization based on this code.
| Database | Precision (%) | Recall (%) | F-measure (%) |
|---|---|---|---|
| ICDAR 2015(val) | 74.61 | 80.93 | 77.64 |
Train
If you want to train the model, you should provide the dataset path, in the dataset path, a separate gt text file should be provided for each image, and make sure that gt text and image file have the same names.
Then run train.py like:
python train.py --gpu_list=0 --input_size=512 --batch_size_per_gpu=8 --checkpoint_path=./resnet_v1_50/ \
--training_data_path=./data/ocr/icdar2015/
If you have more than one gpu, you can pass gpu ids to gpu_list(like --gpu_list=0,1,2,3)
Note:
- right now , only support icdar2017 data format input, like (116,1179,206,1179,206,1207,116,1207,"###"), but you can modify data_provider.py to support polygon format input
- Already support polygon shrink by using pyclipper module
- this re-implementation is just for fun, but I'll continue to improve this code.
- re-implementation pse algorithm by using c++ (if you use python2, just run it, if python3, please replace python-config with python3-config in makefile)
Test
run eval.py like:
python eval.py --test_data_path=./tmp/images/ --gpu_list=0 --checkpoint_path=./resnet_v1_50/ \
--output_dir=./tmp/
a text file and result image will be then written to the output path.
Examples
About issues
If you encounter any issue check issues first, or you can open a new issue.
Reference
- http://download.tensorflow.org/models/resnet_v1_50_2016_08_28.tar.gz
- https://github.com/CharlesShang/FastMaskRCNN
- https://github.com/whai362/PSENet/issues/15
- https://github.com/argman/EAST
Acknowledge
@rkshuai found a bug about concat features in model.py.
If this repository helps you，please star it. Thanks.
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].