-
数据格式如下:
![image](https://user-images.githubusercontent.com/28289146/30376283-3f6979ac-98be-11e7-885f-ba11ccc8bc5f.png)
![image](https://user-images.githubusercontent.com/28289146/30376294-4972ba8a…
-
Hi, I meet the following error when running on Windows 10 Command Prompt and x64 Native Tools Command Prompt for VS2017. I followed the pytorch faster-rcnn to perform visualization on the trained mode…
bk-00 updated
6 years ago
-
I am really confused about that the size of the score_map and geo_map are as same as the input image. However, pool5, which is 1/32 of the size of the input image, just is unsampled 3 times, which mea…
-
Binarization is a key early step for many processes so improvements here can have a large impact on an end result. Otsu is a good all around algorithm and very fast but there are [many improvements in…
-
Hey,
I have already download text_recognition_model.zip in this [web](https://bartzi.de/research/stn-ocr#text-recognition) .
It can be successfully implemented .
But it is only to recognize a …
-
I run the `python3 eval.py`, or `python3 run_demo_server.py`, it will run `from icdar import restore_rectangle`. However, the terminal show that `1000 training images in ./data/train/` and then it may…
-
![image](https://user-images.githubusercontent.com/28289146/30422212-ddea9a9e-9971-11e7-8eea-57665618d338.png)
-
Time series computed using _CitationRatios()_ function have different size (A->B; B->A series). Thus, it is impossible to compute later _GrangerTest_ (it requires equal in size series). There are 401 …
-
I used the command
python eval_seglink.py --checkpoint_path=./seglink/model.ckpt-136750 --dataset_name=icdar2015 --dataset_split_name=test --dataset_dir=./tf_records
to evaluate the model provi…
-
I have trained EAST model with my own generated text image and text coordinate, the text rectangle generated by my own trained model sometimes is shorter than the text, but the generated rec…