Xiaoyan comments

Results 12 comments of


                                            Xiaoyan

exuse me！！！！i have question,my data is image which full of long words,just like books,but when i train model by my data,it can not detect all the words in one line.maybe it can only detect half of a line and my input_size is 768 and text_scale is 768.could you tell me where i maybe wrong?thank you so much

CTPN works badly on oriented images. Does anyone make east to work on horizontal long text images?

If there any method to merge oriented small box into oriented line?

@loulan-D Currently, I merge box following CTPN poly.

If there any method to merge oriented small box into oriented line?

@loulan-D I have re-trained east by my own dataset. But still face the same problem. The long text images are bad. But EAST detect text angle is very accurate. For...

If there any method to merge oriented small box into oriented line?

@loulan-D Maybe east works bad on long-text for this: https://github.com/argman/EAST/issues/94

If there any method to merge oriented small box into oriented line?

@loulan-D If most of your images are horizontal boxes，maybe you should use CTPN. ![image](https://user-images.githubusercontent.com/17979833/40223324-373efeb4-5ab5-11e8-917e-5842f05c19e9.png)

If there any method to merge oriented small box into oriented line?

@loulan-D I use the default score threshold and nms threshold in this https://github.com/argman/EAST/blob/master/eval.py#L69. But I get a better result when set `max_side_len = 512` in https://github.com/argman/EAST/blob/master/eval.py#L39

How can I use this model to train on other language such as Chinese?

Do you fix this?

This convolution is not supported by cudnn, MXNET convolution is applied

I have met this problem. Anyone fix it ?

The third parameter 'rectangle'?

@YangZeyu95 I also met the problems that slower than before. Do you find the reason?

detecting chinese in images

@chajbn I also met the same problem. When training my own Chinese scene text image, the final loss is about 1.5. And do you have any suggestion?