You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am a postgraduate of BUPT in China. First, I am sorry to create a new issue in ICDAR2019_cTDaR to discuss a problem with the ctdar_measurement_tool,because I can't find the 'issue' button on the tool's page.
when I tested and verified the function of ctdar_measurement_tool in table recognition task on the official website of ICDAR 2019 cTDaR, I found there are something wrong with it. General speaking, the question is that I found the result got by the tool was wrong when I increased the number of blank cells between two content cells. You missed to calculate the number of the blank cells in the evaluation tool, which is obviously not consistent with your method description. At the same time, ignoring changes in the number of blank cells isn’t a correct operation in the table recognition task. In order to ensure the correctness of my experiment, I can explain my verification process in detail.
In order to explain the problem in detail, can you give me an email address?
The text was updated successfully, but these errors were encountered:
I am a postgraduate of BUPT in China. First, I am sorry to create a new issue in ICDAR2019_cTDaR to discuss a problem with the ctdar_measurement_tool,because I can't find the 'issue' button on the tool's page.
when I tested and verified the function of ctdar_measurement_tool in table recognition task on the official website of ICDAR 2019 cTDaR, I found there are something wrong with it. General speaking, the question is that I found the result got by the tool was wrong when I increased the number of blank cells between two content cells. You missed to calculate the number of the blank cells in the evaluation tool, which is obviously not consistent with your method description. At the same time, ignoring changes in the number of blank cells isn’t a correct operation in the table recognition task. In order to ensure the correctness of my experiment, I can explain my verification process in detail.
In order to explain the problem in detail, can you give me an email address?
Hi, Is the home page of this competition still available? I tried a few times, but it always told me that the domain name of the site can not be parsed.
@cndplab-founder
Hi cndplab-founder,
I am a postgraduate of BUPT in China. First, I am sorry to create a new issue in ICDAR2019_cTDaR to discuss a problem with the ctdar_measurement_tool,because I can't find the 'issue' button on the tool's page.
when I tested and verified the function of ctdar_measurement_tool in table recognition task on the official website of ICDAR 2019 cTDaR, I found there are something wrong with it. General speaking, the question is that I found the result got by the tool was wrong when I increased the number of blank cells between two content cells. You missed to calculate the number of the blank cells in the evaluation tool, which is obviously not consistent with your method description. At the same time, ignoring changes in the number of blank cells isn’t a correct operation in the table recognition task. In order to ensure the correctness of my experiment, I can explain my verification process in detail.
In order to explain the problem in detail, can you give me an email address?
The text was updated successfully, but these errors were encountered: