TextZoom
TextZoom copied to clipboard
A super-resolution dataset of paired LR-HR scene text images
I would like to remove "TypeError: cannot pickle 'Environment' object" error by making changes in the code .
    每个图中从上到下依次是LR、SR、HR,可以看到SR图像的背景颜色很奇怪,我不知道这是不是导致训练中image loss一直是0的原因,希望能得到解答,谢谢~~
Hi, I tried to execute demo version. python3 main.py --demo --demo_dir='dataset' --resume='model_best.pth' --STN --mask But there is one issue.  So, first I found the same issue in this github...
Hi everyone, I really appreciate the work done both by the proposed paper and within code. I'm trying to reproduce the performance showed in the paper for a Thesis work...
@JasonBoy1 can you release pre-trained models?
用训练好的textzoom的权重 测试的时候达到了70%以上的准确度 但是在跑demo.py时 依旧测试textzoom数据时 输出的预测结果 lr---》sr lr的预测结果准确很多 sr的结果基本上都是错的 都是the 这种单词 是怎么回事呀 求助!
I can start test successfully using Moran and CRNN. However,when I tried to use Aster to test,a problem occured. Have you ever met this problem before? Please help me. 