MaybeShewill-CV/CRNN_Tensorflow
Convolutional Recurrent Neural Networks(CRNN) for Scene Text Recognition
PythonMIT
Pinned issues
Issues
- 29
txt文件与图像间的关系
#354 opened by wandaoyi - 7
中文模型转tensorflow serving后识别速度超慢
#366 opened by guanshuicheng - 2
could you please tell me the solution for this?iam getting this:SyntaxError: Non-UTF-8 code starting with '\xff' in file test_02.jpg on line 1, but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details
#381 opened by santhubaby - 8
- 14
- 1
关于中文数据集转为tfrecords
#384 opened by zhiweiguo - 4
Any suggestion for speeding up the prediction part?
#377 opened by tabsun - 4
crnn_chinese,Restoring from checkpoint failed
#386 opened by wzqy56789 - 12
做中文预测,代码需要有点小改进,不然,result会为空
#353 opened by wandaoyi - 6
多个连续数字串的识别,如何提高准确率?
#376 opened by BigPandaCPU - 1
- 0
- 5
关于准确率
#380 opened by ballshapesdsd - 2
识别一行汉字的数目有限制吗?
#379 opened by ballshapesdsd - 2
- 0
CPU_MULTI_PROCESS_NUMS
#375 opened by Ryansanity - 1
Upgrade to tensorflow 2.0
#369 opened by sachaarbonel - 4
- 1
The link of Chinese pre-training model updated on March 27 cannot be opened. Can you resend it again? Thank you.
#373 opened by yanglinxiabuaaa - 8
Support tflite
#365 opened by sachaarbonel - 6
Error calling crnn_model
#371 opened by SimonWang00 - 4
Excuse me, can we support binary image training
#368 opened by SimonWang00 - 0
从https://www.dropbox.com/sh/z22xsn4byddalv3/AAAiIxAHJKbqy44M73ow5znSa?dl=0&lst=下载的中文的模型,识别出来都是空。
#370 opened by ZhangZoffy - 2
Model does not converge on larger dataset
#367 opened by guddulrk - 5
用自己的数据运行train以后 程序被killed 并非代码报错会是什么原因呢
#348 opened by GreenTree1990 - 14
- 2
请问作者,准确率是怎么计算的啊
#364 opened by yuxiaomu - 1
生成的tfrecords为什么这么大?
#362 opened by LJXLJXLJX - 1
Confidence level for each word
#357 opened by eravallirao - 5
将配置文件中的图片高度项改成48,训练的时候报错是为什么呢?是网络对输入图片的高度有限制吗?谢谢
#355 opened by Leedorado - 1
crnn_synth90k的模型是不是不能识别数字?
#360 opened by ZhangZoffy - 4
我用了90k的那个模型报错
#359 opened by ZhangZoffy - 3
Get position of definied symbols
#356 opened by oleshkoo - 0
use crnn_synth90k model error
#358 opened by ZhangZoffy - 14
请问,为什么我用自己的数据集训练出来的准确率能有百分之八九十,但测试的准确率总是百分之二十几啊
#338 opened by yuxiaomu - 3
- 6
Retraining for line with more than two words
#351 opened by eravallirao - 4
用自己的中英文数据集训练,只能识别字母与数字,所有的汉字以及标点全都识别成空。global_config中相关项已经做了更改,为什么会出现这种情况呢?thx~
#345 opened by Leedorado - 2
Assign requires shapes of both tensors to match. lhs shape= [512,37] rhs shape= [512,5825]
#352 opened by XP-CHRC - 1
- 4
能把权重发给我吗?您的那个网站打不开
#347 opened by S130111 - 1
Is it possible to retrain the model with my own images from the checkpoint weights you made available?
#344 opened by vinicius-mariano - 1
- 0
关于相邻的重复字符的识别
#346 opened by cosimo17 - 5
tf-gpu版本为1.12,训练的时候报错Failed to get convolution algorithm. This is probably because cuDNN failed to initialize,请问这种问题应该如何解决?谢谢~~
#343 opened by Leedorado - 2
TF Records
#337 opened by vinicius-mariano - 6
请问lexicon.txt的行数是从第零行还是第一行开始计数?
#340 opened by Leedorado - 2
请问crnn能识别未见过的词条组合吗?
#342 opened by mymynew - 1
can not download the pretained checkpoint model
#339 opened by 13438960761 - 1
我在运行到一半时出现了下面的问题,标签长度为零,请问这是什么意思,是对应某一行没有标签吗,我该怎么排查
#336 opened by yuxiaomu