Real Time Object Detection Details
Upload an image and the API will return all the objects (including people) in the image, their confidence and the coordinates of their bounding boxes.
It returns a simple JSON object containing the most essential informations.
Example:
{
“labels” :
[
{’ coordinates’:
{‘top-left’: (2564, 1524), ‘top-right’: (2671, 1524), ‘bottom-left’: (2564, 1972), ‘bottom-right’: (2671, 1972)},
‘label’: ‘person’,
‘confidence’: 66.49194359779358} ,
{‘coordinates’: {‘top-left’: (2753, 1572), ‘top-right’: (2890, 1572), ‘bottom-left’: (2753, 1910), ‘bottom-right’: (2890, 1910)}, ‘
label’: ‘dog’,
‘confidence’: 71.51359915733337 }
, ……
, ……
]
}
This API uses cutting edge deep convolutional neural network architecture trained over millions of examples.