Closed zobeirraisi closed 2 years ago
In the paper, we mentioned that "We discard images with widths shorter than 32 pixels as they are too blurry, and obtain 4.1M word images in total. "
For the 3Mx3 experiment, I randomly select 3M from this 4.1M.
In the paper, we mentioned that "We discard images with widths shorter than 32 pixels as they are too blurry, and obtain 4.1M word images in total. "
For the 3Mx3 experiment, I randomly select 3M from this 4.1M.
Is it possible to share those extracted images?
In the paper, we mentioned that "We discard images with widths shorter than 32 pixels as they are too blurry, and obtain 4.1M word images in total. " For the 3Mx3 experiment, I randomly select 3M from this 4.1M.
Is it possible to share those extracted images?
Sorry, I didn't keep those data after I left the company.
I recommend that you filter out text that are too small.
If it's possible, you can set the FontSize
flag and generate a new dataset with larger text.
How you select the 3M word image from the 150 GB dataset?