-
hi, @MhLiao
I prepared my own dataset, there were 500 720*1280 photos, I marked these pictures, and then trained. I used 6 NVIDIA K80 GPUs, but the training speed is very slow, 10 iterations take an…
-
Hi, great repo !
Judging from the code the implementation is not complete, for example the [Recognizer](https://github.com/jiangxiluning/FOTS.PyTorch/blob/79e226b763ad14255bcc2b44d9a25a7941c7a29b/m…
-
## What
- 규민님 요청에 따라 poly_ill.json 학습 시작
- 실습 annotation dataset, polygon illegibility True 진행 후 학습
## Why?
- 실험 및 성능 비교
## Todo
## 작업 내용
1. git develop에서 최신 버전을 pull
2. poly_ill.json 서버…
-
@zhang0jhon
博主您好，首先特别感谢您做的工作，您开源的模型，效果确实很好。
我想尝试复现一下训练流程，但遇到如下3个问题：
1）速度特别慢，我只用了LSVT的数据，一个epoch都要大约6个小时
2）我尝试用多卡训练，但与单卡速度相当，我用的2080ti的卡
3）我测试了30个epoch后的效果，识别精度很差
想请教下：
1）模型训练，需要多少个epoch才合适…
-
Hi, when I run eval.py in anaconda-project environment, after the program finished, a "Segmentation fault (core dumped)" may occured randomly .
operation history:
```
$ docker run -t -i conti…
-
hi, I found that it is not so easy to split the Gaussian distribution map.
can you provide details of the watered algorithm?
For example, the binarization method used here?…
-
参考：http://blog.csdn.net/peaceinmind/article/details/51387367
# ICDAR数据集
ICDAR大赛组委会自2003年起公布了自然场景文本数据库，直到2015年该数据库已包括文本定位数据库、文本分割数据库、单词识别数据库、端对端识别数据库。图像都是是24位彩色图像，大小是1024*768像素。
# SCUT-FORU-DB数据库。
…
-
https://arxiv.org/pdf/2203.02378.pdf @ ACM 2022
-
Hi,
I just don't know how to start with the text recognition for another language.
I have tried changing 'codec' variable (add more non-English characters), and I think I should change 'caffe.Spatia…
ghost updated
6 years ago
-
**Reproduction**
1. What command or script did you run?
```none
python tools/test.py configs/textdet/dbnet/dbnet_r50dcnv2_fpnc_1200e_totaltext.py experiments/dbnet_r50_totaltext/epoch_25.pth --…
ahsdx updated
2 years ago