I am facing issue with google cloud vision api response. there is no issue with proper text image. the problem is with curved shape text written image. i attached for reference. when i try to detect text from image that words are shuffling and its not coming properly.
the json i am getting through response is having different coordinate and i could not arrange it properly as shown in the image.
"boundingPoly": {
"vertices": [
{
"x": 395,
"y": 21
},
{
"x": 421,
"y": 18
},
{
"x": 423,
"y": 39
},
{
"x": 398,
"y": 42
}
]
},
"description": "on"
},
{
"boundingPoly": {
"vertices": [
{
"x": 442,
"y": 16
},
{
"x": 482,
"y": 11
},
{
"x": 484,
"y": 31
},
{
"x": 444,
"y": 36
}
]
},
"description": "the"
},
{
"boundingPoly": {
"vertices": [
{
"x": 263,
"y": 22
},
{
"x": 297,
"y": 18
},
{
"x": 300,
"y": 45
},
{
"x": 266,
"y": 49
}
]
},
"description": "all"
},
{
"boundingPoly": {
"vertices": [
{
"x": 316,
"y": 15
},
{
"x": 382,
"y": 7
},
{
"x": 385,
"y": 35
},
{
"x": 319,
"y": 43
}
]
},
"description": "items"
},
{
"boundingPoly": {
"vertices": [
{
"x": 330,
"y": 57
},
{
"x": 399,
"y": 61
},
{
"x": 397,
"y": 99
},
{
"x": 328,
"y": 95
}
]
},
"description": "have"
},
{
"boundingPoly": {
"vertices": [
{
"x": 420,
"y": 63
},
{
"x": 501,
"y": 68
},
{
"x": 499,
"y": 105
},
{
"x": 418,
"y": 100
}
]
},
"description": "impact"
},
the json is very lengthy so i added very few
i have tried here - >google vision api demo
google vision api demo api result
the expected result should be same as the image shown
i am struggling with this past few days. all your support and help will be appreciated.