-
您好，麻烦请问pytorch版的ctpn性能如何，结合crnn，识别一张图片大概需要多长时间，有用到GPU还是在CPU上计算的？
我在用CRNN的时候，发现CPU版的耗时很长，一张身份证图片，crnn得十几秒(32core E5 v3的cpu)，不知道是什么原因。
您在实际使用的时候，FPS大概能达到多少？
O(∩_∩)O谢谢
-
def __getitem__(self, index):
label = [int(x) for x in keys]
请问label是要识别的str，怎么转成int呢？
-
使用ncnn_project下的crnn显示LSTM层不支持。
./Text_recognition ../models/crnn_lite_lstm_v2.bin ../models/crnn_lite_lstm_v2.param test.jpg
输出：
输入尺寸 (277, 32, 1)
layer LSTM not exists or registered
段错误 (核心已转储)…
-
训练时候运行crnn_main.py
Traceback (most recent call last):
File "crnn_main.py", line 310, in
cost = trainBatch(crnn, criterion, optimizer)
File "crnn_main.py", line 257, in trainBatch
…
-
Recently TextBoxes ++ https://github.com/MhLiao/TextBoxes_plusplus , uploaded their paper and in their implementation for text recognition part they are using CRNN and they also referred to CRNN imple…
-
Hi--
I was wondering what the difference was between the pre-trained crnn_kurapan.h5 and crnn_kurapan_notop.h5 recognizer models was?
And what were they trained on?
Thanks for the fantastic l…
-
As far as I know, the only way to use doctr on gpu is by adding `to(device)` like this
```python
from doctr.models import ocr_predictor, from_hub
reco_arch = from_hub('tilman-rassy/doctr-crnn…
-
### Describe the issue
when i test OCR CRNN model with onnxruntime-gpu-1.10 ,it gets different recognise results, that's amazing, mode weights is not stable
### To reproduce
when i test OCR CRNN mo…
-
我用create_dataset.py处理了自己的训练数据，然后放入CRNN中训练，出现以下错误：
/crnn_net/dataset.py:95: UserWarning: torch.range is deprecated in favor of torch.arange and will be removed in 0.5. Note that arange generates value…
-
## 🐛 Bug
## To Reproduce
Steps to reproduce the behavior:
I trained the model on Pytorch and deployed it on libtorch, which I tried in both versions 1.1 and 1.4. They all run successfully…