-
i have issues when i trained model with 13k sample
Traceback (most recent call last):
File "train.py", line 319, in
train(args)
File "train.py", line 143, in train
for i, (images,…
-
Does anybody have working comiled version?
All compiliations is very slow and crash windows
-
Hello!
Here a few exaimples how to use OpenCV to scan / transform and prepare scanned docuemnts. Results are much better then current scantailor release.
Would be nice to have opencv as part of …
axet updated
10 months ago
-
希望矫正之后的图片保持与原图差不多的分辨率，如何调整参数？矫正后的图片不失真。
另外，目前发现矫正后的文字发生变形，有时变得扁平，有时变瘦长，能否做到不变形？谢谢！
-
I'm not sure whether this is the right place to ask as `sbb-textline-detector` itself worked perfectly in our OCR-D workflows and the produced segmentation results look good as well but running any re…
-
Hi,
first I want to thank you for your work.
The algorithm works fine for images, where all the document can be seen and is flat.
However, I have a case of a document, where the dimensions of t…
-
Thanks for sharing the code and dataset. The encoder-only architecture makes DDCP faster and lighter than other methods, I really like the idea. I try to reimplement the paper, however, some training …
-
Could you tell which paper do you refer in this way? Thanks@gwxie
-
@zvezdochiot решил отдельно завести для варнингов под юниксы тикет, чтобы не загромождать под винду
Вот debian12 x64 с `-D CMAKE_BUILD_TYPE=Release`, boost я в список не включал. Завтра попробую эту…
-
Hi @yoyoberenguer
Great to see your code fine . In you solution you have taken single flat image and converted to 360 degree fish eye image.
Do you have any any idea/algo which revere the output fr…