Hive's text recognition model detects and transcribes each word in an image. It can also returned semantically grouped and ordered text blocks in their natural reading order for words that are grouped closely together.
Hive’s text recognition response format is an instantiation of the detection response format that outputs a confidence score for each detected object (word), an additional confidence score for the transcription of the characters in the detected words, as well as a separate field containing the aforementioned block text.
Word detections are returned in the bounding_poly list.
{
"id": "a63f4890-5b06-11ed-8d29-81ceb2961e0b",
"code": 200,
"project_id": 41563,
"user_id": 3121654,
"created_on": "2022-11-02T23:32:44.603Z",
"status": [
{
"status": {
"code": "0",
"message": "SUCCESS"
},
"response": {
"input": {
"id": "a63f4890-5b06-11ed-8d29-81ceb2961e0b",
"charge": 0.003,
"model": "...",
"model_version": 1,
"model_type": "BOUNDING_BOX",
"created_on": "2022-11-02T23:32:43.545Z",
"media": {
"url": null,
"filename": "SanFranciscoNeighborhoods.jpeg",
"type": "PHOTO",
"mime_type": "jpeg",
"mimetype": "image/jpeg",
"width": 1600,
"height": 1314,
"num_frames": 1,
"duration": 0
},
"user_id": 3121654,
"project_id": 41563
},
"output": [
{
"block_text": "San Francisco NORTH california WATERSRONT MARINA HEBON TI เNนามL RUSSIAN PRESIDIO HOLLOW HILL BEACH COW FINANCIAL DISTRICT PACIFIC NOB HILL BARBARY COAST PRESIDIO HEIGHTS MA HEIGHTS HEIGHTS DOWNTOWN CLIFF STREET LOWER PACIFIC LINCOLN LAKE PARK ИАСКСИ PARK CVIC VAN LAUREL HERHTS TENDERLOIN BUENA SOUTH BEACH ANZA WESTERN CENTER NEMA CENTRAL INNER VISTA ADDITION RICHMOND LONE OUTER RICHMOND MOUNTAIN RICHMOND NORTH ALAMO PANHANDLE SQUARE SOUTH OF HAYES MARKET NAIGHT ANHBURY VALLEY MISSION GOLDEN GATE PARK NEN 0B VOSAN DAXE BAY mESJmS SAVLLE LENA VALIAT INNER PANASSIN CORCINA MISSION DOLORES 코드21 HEKIHTS SUNSET OUTER CENTRAL HLKIKA VALLEY FOREST DOLDIES HEIGHTS POTRERO HILL SUNSET SUNSET GOLDEN KNOLLS TWIN PEAKS INNER MISSION CENTRAL WATERFRONT DOGPATCH LANE TERRACE NOE HEKMITS FOREST VALLEY OUTER ... PARKSIDE PARKSIDE INNER PARKSIDE WEST CN PL. PORTAL MIRALOMA HEIGHTS BERNAL HUNTERS HEIGHTS SANT PARK PARK PINE LAKE PARK WEE GLEN MER-E MANOIR SERRALT Bな版I SUNNYSIDE SILVER MOUNT BAYVIEW SESON LAKESIDE MANE WESTWOOD PARK MISSION PORTOLA POINT LAKE INGUESIDE TERRACE SHORE STONESTOWN HEIGHTS MERCED INGLESIDE OCEANVIEW EXCELSIOR MEA사엘 HEIGHTS MISSION VISITACION INGLESIDE HEIGHTS CROCKER VALLEY CANDLESTICK POINT AMAZON AITILL NSITRLENI",
"bounding_poly": [
{
"classes": [
{
"class": "San"
}
],
"dimensions": {
"left": 1130.8434218203035,
"right": 1229.3265023337224,
"top": 54.65066586841236,
"bottom": 96.17978518659417
},
"vertices": [
{
"x": 1130.8434218203035,
"y": 96.17978518659417
},
{
"x": 1130.8434218203035,
"y": 54.65066586841236
},
{
"x": 1229.3265023337224,
"y": 54.65066586841236
},
{
"x": 1229.3265023337224,
"y": 96.17978518659417
}
],
"meta": {
"score": 0.9997242093086243,
"label": "text"
}
},
{
"classes": [
{
"class": "Francisco"
}
],
"dimensions": {
"left": 1265.0711967619604,
"right": 1545.1178712077012,
"top": 55.93030897053805,
"bottom": 96.05957033417442
},
"vertices": [
{
"x": 1265.0711967619604,
"y": 96.05957033417442
},
{
"x": 1265.0711967619604,
"y": 55.93030897053806
},
{
"x": 1545.1178712077012,
"y": 55.93030897053805
},
{
"x": 1545.1178712077012,
"y": 96.05957033417442
}
],
"meta": {
"score": 0.9998077750205994,
"label": "text"
}
},
{
"classes": [
{
"class": "NORTH"
}
],
"dimensions": {
"left": 944.3767320595099,
"right": 987.3172221411902,
"top": 133.31049624356356,
"bottom": 144.5093598799272
},
"vertices": [
{
"x": 944.3767320595099,
"y": 144.5093598799272
},
{
"x": 944.3767320595099,
"y": 133.31049624356356
},
{
"x": 987.3172221411902,
"y": 133.31049624356356
},
{
"x": 987.3172221411902,
"y": 144.5093598799272
}
],
"meta": {
"score": 0.9777765274047852,
"label": "text"
}
},
{
"classes": [
{
"class": "california"
}
],
"dimensions": {
"left": 1222.0575499562426,
"right": 1457.296756490665,
"top": 103.35988140106201,
"bottom": 149.555193901062
},
"vertices": [
{
"x": 1222.0575499562426,
"y": 149.555193901062
},
{
"x": 1222.0575499562426,
"y": 103.35988140106201
},
{
"x": 1457.296756490665,
"y": 103.35988140106201
},
{
"x": 1457.296756490665,
"y": 149.555193901062
}
],
"meta": {
"score": 0.9824895858764648,
"label": "text"
}
},
{
"classes": [
{
"class": "NSITRLENI"
}
],
"dimensions": {
"left": 1031.6242160151692,
"right": 1096.5016955950991,
"top": 1220.854650150646,
"bottom": 1229.7204171961005
},
"vertices": [
{
"x": 1031.6242160151692,
"y": 1229.7204171961005
},
{
"x": 1031.6242160151692,
"y": 1220.854650150646
},
{
"x": 1096.5016955950991,
"y": 1220.854650150646
},
{
"x": 1096.5016955950991,
"y": 1229.7204171961005
}
],
"meta": {
"score": 0.9256755709648132,
"label": "text"
}
}
],
"time": 0
}
]
}
}
],
"from_cache": false
}
The additional fields to the detection response format that are unique to the ocr models are:
| Name | Description |
|---|---|
| classes.0.class | Contains the transcribed characters for the detected word. |
| classes.0.score | Contains the confidence score for the transcribed word. |
| meta.score | Contains the confidence score for the detected word — irrespective of the transcription of that word. |
Hive only returns their high confidence predictions to end-users. The scores are provided as additional metadata to users, but users do not need to apply any thresholds or discard predictions to obtain accurate model results.