search

xieenze / TextSR

87 stars 16 forks source link

When can the code be released? #5

Closed kunlaotou closed 4 years ago

kunlaotou commented 5 years ago

I read your paper and I feel that your paper is great. Since I am a newbie, I went to read SRGAN first and did experiments. When can your code be released?thx. ^_^

WenjiaWang0312 commented 5 years ago

Dear kunlaotou: We think our work is still too weak. Now we are doing a substantial modification to our work and we will postpone the releasing for one or two months. Our paper would be re-wrote as well.

kunlaotou commented 5 years ago

Dear kunlaotou: We think our work is still too weak. Now we are doing a substantial modification to our work and we will postpone the releasing for one or two months. Our paper would be re-wrote as well.

铁汁，那祝你们顺利，我再等等~

qiu-pinggaizi commented 5 years ago

大兄弟 你们这个实验结果看起来非常棒呀！ 什么时候可以开源代码呢

WenjiaWang0312 commented 5 years ago

大兄弟 你们这个实验结果看起来非常棒呀！ 什么时候可以开源代码呢

重做中

qiu-pinggaizi commented 5 years ago

读完你的paper后 我有两个问题想请教一下：1、在github中，你说使用的图像是大于128*32的图像；但在paper中是smaller，这和前者是冲突的呀，所以想要确认一下下。

2、在获取图像中text中，你使用了crop，想请教一下具体怎么处理的；我这边只是想对text做处理，所以想只提取图像中的text信息，如果可以希望可以有相关代码，感谢。

WenjiaWang0312 commented 5 years ago

读完你的paper后 我有两个问题想请教一下：1、在github中，你说使用的图像是大于128*32的图像；但在paper中是smaller，这和前者是冲突的呀，所以想要确认一下下。 2、在获取图像中text中，你使用了crop，想请教一下具体怎么处理的；我这边只是想对text做处理，所以想只提取图像中的text信息，如果可以希望可以有相关代码，感谢。

  1. The 'bigger' images are train set, while the 'smaller' images are used as test set.
  2. I have uploaded a script: 'crop_800k.py'. You can use this to crop the Syntxt_800K dataset. Remember to change the params.
qiu-pinggaizi commented 5 years ago

OK I have downloaded the .py file, I will have a try, thank you.

发自我的iPhone

------------------ Original ------------------ From: Wenjia Wang <notifications@github.com> Date: 周二,11月 5,2019 20:09 To: xieenze/TextSR <TextSR@noreply.github.com> Cc: qiu-pinggaizi <1135771327@qq.com>, Comment <comment@noreply.github.com> Subject: Re: [xieenze/TextSR] When can the code be released? (#5)

读完你的paper后 我有两个问题想请教一下：1、在github中，你说使用的图像是大于128*32的图像；但在paper中是smaller，这和前者是冲突的呀，所以想要确认一下下。 2、在获取图像中text中，你使用了crop，想请教一下具体怎么处理的；我这边只是想对text做处理，所以想只提取图像中的text信息，如果可以希望可以有相关代码，感谢。

The 'bigger' images are train set, while the 'smaller' images are used as test set.

I have uploaded a script: 'crop_800k.py'. You can use this to crop the Syntxt_800K dataset. Remember to change the params.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

RamanHacks commented 4 years ago

Hi Jason, impressive work! Wanted to know if you will release the code soon? Was working on a similar project that's why I was curious! Thanks!

WenjiaWang0312 commented 4 years ago

Hi Jason, impressive work! Wanted to know if you will release the code soon? Was working on a similar project that's why I was curious! Thanks! New version😄 https://arxiv.org/abs/2005.03341