The proprietary JSON annotation format used by the Cogniac computer vision platform.
Below, learn the structure of Cogniac.
{
"frame": 0,
"filename": "./PATH_TO_FILE.ext",
"object_name_1": [
{
"class_name_1": {
"x0": 1334,
"x1": 1657,
"y0": 242,
"y1": 553,
"probability": 0.8974609375
},
"class_name_2": [],
"class_name_3": []
},
{
"class_name_1": {
"x0": 7,
"x1": 201,
"y0": 31,
"y1": 337,
"probability": 0.87353515625
},
"class_name_2": [
{
"text": "476240",
"x0": 137,
"x1": 156,
"y0": 128,
"y1": 206,
"probability": 0.755047082901001
}
],
"class_name_3": []
},
{
"class_name_1": {
"x0": 547,
"x1": 844,
"y0": 157,
"y1": 442,
"probability": 0.7021484375
},
"numbers": [],
"initials": []
}
],
"object_name_2": [],
"object_name_3": [],
"object_name_4": [
{
"text": "36",
"x0": 628,
"x1": 728,
"y0": 571,
"y1": 641,
"probability": 0.9057629108428955
},
{
"text": "37",
"x0": 987,
"x1": 1101,
"y0": 652,
"y1": 730,
"probability": 0.9051821231842041
},
{
"text": "38",
"x0": 1418,
"x1": 1542,
"y0": 757,
"y1": 839,
"probability": 0.9096717238426208
}
]
}
With Roboflow supervision, an open source Python package with utilities for completing computer vision tasks, you can merge and split detections in Cogniac. Read our dedicated guides to learn how to merge and split Cogniac detections.
Below, see model architectures that require data in the Cogniac format when training a new model.
On each page below, you can find links to our guides that show how to plot predictions from the model, and complete other common tasks like detecting small objects with the model.