-
In the test stage, the short sides of the test image are set to 736, 736, 1080, 1600 for Total-Text, CTW1500, ICDAR2015, and ArT, while the long sides are resized to keep the original aspect ratio.
比…
-
新手求助 下载gt_ctw1500 gt_icdar2015链接失效了
-
想请问问学长，evaluate一下ctw1500，为什么指标始终为0呢，明明demo是可以可视化成功的
是框架的问题吗？以下是指令：
python tools/train_net.py -…
-
I ran the total-text and the results were good, so I would like to try the CTW1500 dataset. Could you please provide the config yaml for CTW1500, thanks.
-
https://cloudstor.aarnet.edu.au/plus/s/uoeFl0pCN9BOCN5/download 链接失效了
-
Hi, thanks for the great work.
I have some issues with the evaluation performance on the CTW1500 dataset.
If I evaluate using the code from this github, the performance is the same as reported.
How…
-
Hi, thanks for your solid work.
Could you please provide the dataset link and pre-processing information for CTW-1500 Dataset?
The same is for other datasets.
It will be nice for you to perfect y…
-
An error occurred when I downloaded the evaluation compressed package of the CTW1500 data set. The download address has expired. Where can I download this zip file?
wget -O gt_ctw1500.zip https://c…
-
感谢作者的工作和代码，以下是个人的复现过程：
* 复现的模型是：Res50-1s
* 训练数据：CTW1500的1000张训练数据，**未使用其他额外数据**
* 训练启动脚本：`python3 -u train_textBPN_author.py --exp_name Ctw1500 --net resnet50 --scale 1 --max_epoch 660 --batch_size…
-
求，非常感谢