PLEASE share the exact guideline on how to get the 8 coordinates x1,y1,x2,y2,x3,y3,x4,y4 for any image text which has to passed into split_label.py ? INORDER TO GET THE 4 coordinates for preparing custom dataset ?

[AjitAntony]

Query1. How the coordinates x1,y1,x2,y2,x3,y3,x4,y4 of the sample image gt_img_859.txt was collected? inorder to feed into split_label.py to get the correct coordinates of bounding box ?

1192,1862,2424,1895,2405,2428,1185,2366,english,8512 846,1280,1092,1379,1088,1478,831,1390,english,### 299,1123,585,1222,536,1370,240,1291,english,###

when i opened paint tool in windows 10 and checked the coordinates of this image none of the coordinates were matching the boundaries of the 3 text line .

x1,x2,y1,y2,x3,y3,x4,y4 1192,1862,2424,1895,2405,2428,1185,2366,english,8512

How to get the 8 coordinates for my custom image ?

Query2: i followed the instruction shared in the prepare your own dataset

I ran split_label.py with the same image shared in example and with same annotation file gt_img_859.txt . I was was expecting the output of final annotation with 4 coordinates to be same with the annotaion values sahre in this link img_859.txt BUT IT WAS COMPLETELY DIFFERENT

.PLEASE share the exact guideline on how to get the 8 coordinates for any image text which has to passed into split_label.py ?

the final annotations and its coordinates that was given as output after running split_label.py is below 522,819,527,1042 528,819,543,1042 544,820,559,1043 560,820,575,1044 576,821,591,1045 592,821,607,1045 608,821,623,1046 624,822,639,1047 640,822,655,1048 656,823,671,1049 672,823,687,1049 688,824,703,1050 704,824,719,1051 720,824,735,1052 736,825,751,1053 752,825,767,1054 768,826,783,1054 784,826,799,1055 800,827,815,1056 816,827,831,1057 832,827,847,1058 848,828,863,1058 864,828,879,1059 880,829,895,1060 896,829,911,1061 912,830,927,1062 928,830,943,1062 944,830,959,1063 960,831,975,1064 976,831,991,1065 992,832,1007,1066 1008,832,1023,1067 1024,833,1039,1067 1040,833,1055,1068 1056,833,1071,1069 366,560,367,616 368,561,383,617 384,567,399,623 400,574,415,628 416,580,431,633 432,587,447,639 448,593,463,644 464,600,479,650 480,606,495,655 105,485,111,572 112,487,127,574 128,493,143,578 144,498,159,582 160,504,175,586 176,509,191,591 192,515,207,595 208,520,223,599 224,526,239,603 240,531,255,608 256,537,271,612

coordinate values of [img_859.txt] (https://github.com/eragonruan/text-detection-ctpn/blob/banjin-dev/data/readme/img_859.txt) : present in this file is totally different from the output that i got when i ran the split_label.py

238,376,239,479 240,376,255,479 256,377,271,480 272,377,287,481 288,378,303,482 304,378,319,482 320,378,335,483 336,379,351,484 352,379,367,485 368,380,383,486 384,380,399,487 400,381,415,487 416,381,431,488 432,382,447,489 448,382,463,490 464,382,479,491 480,383,495,491 167,257,175,286 176,261,191,289 192,267,207,295 208,274,223,300 48,223,63,265 64,228,79,269 80,234,95,273 96,239,111,278 112,245,127,282

[vanchung1995]

In split_label.py, it use only 8 first numbers. So it is unnecessary to care about it. 8 first number is x1,y1,x2,y2,x3,y3,x4,y4 respect to top left, top right, down right, down left

[lrfighting]

在split_label.py中，它仅使用8个前数字。因此没有必要在意它。 8个第一个数字是x1，y1，x2，y2，x3，y3，x4，y4关于左上，右上，右下，左下

请问split_label.py这个文件运行后的标签文件里的数字具体是什么意思呢

[Mrlei-go]

Query1. How the coordinates x1,y1,x2,y2,x3,y3,x4,y4 of the sample image gt_img_859.txt was collected? inorder to feed into split_label.py to get the correct coordinates of bounding box ?

1192,1862,2424,1895,2405,2428,1185,2366,english,8512 846,1280,1092,1379,1088,1478,831,1390,english,### 299,1123,585,1222,536,1370,240,1291,english,###

when i opened paint tool in windows 10 and checked the coordinates of this image none of the coordinates were matching the boundaries of the 3 text line .

x1,x2,y1,y2,x3,y3,x4,y4 1192,1862,2424,1895,2405,2428,1185,2366,english,8512

How to get the 8 coordinates for my custom image ?

Query2: i followed the instruction shared in the prepare your own dataset

I ran split_label.py with the same image shared in example and with same annotation file gt_img_859.txt . I was was expecting the output of final annotation with 4 coordinates to be same with the annotaion values sahre in this link img_859.txt BUT IT WAS COMPLETELY DIFFERENT

.PLEASE share the exact guideline on how to get the 8 coordinates for any image text which has to passed into split_label.py ?

the final annotations and its coordinates that was given as output after running split_label.py is below 522,819,527,1042 528,819,543,1042 544,820,559,1043 560,820,575,1044 576,821,591,1045 592,821,607,1045 608,821,623,1046 624,822,639,1047 640,822,655,1048 656,823,671,1049 672,823,687,1049 688,824,703,1050 704,824,719,1051 720,824,735,1052 736,825,751,1053 752,825,767,1054 768,826,783,1054 784,826,799,1055 800,827,815,1056 816,827,831,1057 832,827,847,1058 848,828,863,1058 864,828,879,1059 880,829,895,1060 896,829,911,1061 912,830,927,1062 928,830,943,1062 944,830,959,1063 960,831,975,1064 976,831,991,1065 992,832,1007,1066 1008,832,1023,1067 1024,833,1039,1067 1040,833,1055,1068 1056,833,1071,1069 366,560,367,616 368,561,383,617 384,567,399,623 400,574,415,628 416,580,431,633 432,587,447,639 448,593,463,644 464,600,479,650 480,606,495,655 105,485,111,572 112,487,127,574 128,493,143,578 144,498,159,582 160,504,175,586 176,509,191,591 192,515,207,595 208,520,223,599 224,526,239,603 240,531,255,608 256,537,271,612

coordinate values of [img_859.txt] (https://github.com/eragonruan/text-detection-ctpn/blob/banjin-dev/data/readme/img_859.txt) : present in this file is totally different from the output that i got when i ran the split_label.py

238,376,239,479 240,376,255,479 256,377,271,480 272,377,287,481 288,378,303,482 304,378,319,482 320,378,335,483 336,379,351,484 352,379,367,485 368,380,383,486 384,380,399,487 400,381,415,487 416,381,431,488 432,382,447,489 448,382,463,490 464,382,479,491 480,383,495,491 167,257,175,286 176,261,191,289 192,267,207,295 208,274,223,300 48,223,63,265 64,228,79,269 80,234,95,273 96,239,111,278 112,245,127,282

HAVE YOU SOLVED IT?

[dahalsweekar]

@Mrlei-go Go to his repo and follow the instructions.

https://github.com/ajinkya933/text-detection-ctpn