Belval / Crnn
Licence: mit
A TensorFlow implementation of https://github.com/bgshih/crnn
Programming Languages
python 139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Crnn
AESRC2020a deep accent recognition network
Stars: ✭ 35 (-87.8%)
Mutual labels: ctc, crnn
CRNN.tf2Convolutional Recurrent Neural Network(CRNN) for End-to-End Text Recognition - TensorFlow 2
Stars: ✭ 131 (-54.36%)
Mutual labels: ctc, crnn
SightseqComputer vision tools for fairseq, containing PyTorch implementation of text recognition and object detection
Stars: ✭ 116 (-59.58%)
Mutual labels: crnn, ctc
crnn一个基于TensorFlow2的CRNN项目
Stars: ✭ 47 (-83.62%)
Mutual labels: crnn
BankCard-RecognizerIdentifying numbers from bankcard, based on Deep Learning with Keras [China Software Cup 2019]
Stars: ✭ 74 (-74.22%)
Mutual labels: crnn
CRNN-OCR-liteLightweight CRNN for OCR (including handwritten text) with depthwise separable convolutions and spatial transformer module [keras+tf]
Stars: ✭ 130 (-54.7%)
Mutual labels: crnn
torch-asgAuto Segmentation Criterion (ASG) implemented in pytorch
Stars: ✭ 42 (-85.37%)
Mutual labels: ctc
insightocrMXNet OCR implementation. Including text recognition and detection.
Stars: ✭ 100 (-65.16%)
Mutual labels: crnn
tensorflow-crnntensorflow slim Implementation crnn
Stars: ✭ 17 (-94.08%)
Mutual labels: crnn
crnn.gluonA gluon re-implementation of Convolutional recurrent network in gluon
Stars: ✭ 21 (-92.68%)
Mutual labels: crnn
ctc-asrEnd-to-end trained speech recognition system, based on RNNs and the connectionist temporal classification (CTC) cost function.
Stars: ✭ 112 (-60.98%)
Mutual labels: ctc
rnn benchmarksRNN benchmarks of pytorch, tensorflow and theano
Stars: ✭ 85 (-70.38%)
Mutual labels: ctc
ssd detectorsSSD-based object and text detection with Keras, SSD, DSOD, TextBoxes, SegLink, TextBoxes++, CRNN
Stars: ✭ 295 (+2.79%)
Mutual labels: crnn
Multimodal-Gesture-Recognition-with-LSTMs-and-CTCAn end-to-end system that performs temporal recognition of gesture sequences using speech and skeletal input. The model combines three networks with a CTC output layer that recognises gestures from continuous stream.
Stars: ✭ 25 (-91.29%)
Mutual labels: ctc
crnn.caffecrnn.caffe
Stars: ✭ 76 (-73.52%)
Mutual labels: crnn
Cortex License Plate Reader ClientA client to connect to cortex-provisioned infrastructure on AWS to do license plate identification in real time.
Stars: ✭ 268 (-6.62%)
Mutual labels: crnn
Tf CrnnTensorFlow convolutional recurrent neural network (CRNN) for text recognition
Stars: ✭ 275 (-4.18%)
Mutual labels: crnn
This repository has been archived. Please see all the other (much better) implementations around GitHub
CRNN
A TensorFlow implementation of https://github.com/bgshih/crnn
But what is a CRNN?
It is a Convolutional Recurrent Neural Network that can be used as an OCR
Requirements
- Tensorflow (tested with 1.8)
pip3 install tensorflow
- Scipy
pip3 install scipy
What training data was used?
All training data (200 000 examples) were generated using my other project https://github.com/Belval/TextRecognitionDataGenerator
To do the same, simply install that project with pip (pip install trdg
) and do trdg -c 200000 -w 1 -t 8
. -t
should be your processor thread count.
Pretrained model
Available in CRNN/save. Use python3 run.py -ex ../data/test --test --restore
to test.
Specify charset
You can specify charset to include only numbers python run.py --train -ex ../data/test -it 50000 -cs 0123456789
Results
It works but is a suboptimal solution for OCR in its current form as it makes some mistakes. Do note that I used a bigger char vector than the paper.
For fun, here are a list of words with their prediction:
Ground truth |
Prediction |
Image |
retardates |
retardates |
|
pay-roller |
poy-roler |
|
rhizopodist |
rhizospodist |
|
theriacas |
trenagas |
|
semantically |
semanticaly |
|
dualistic |
duaistic |
|
high-flying |
highi-fling |
|
grossify |
grsity |
|
scutular |
scutular |
|
crispened |
crispened |
|
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].