This model detects all the faces in an image, then takes each detected face as an input to analyze the pose of the head based on the data obtained through a face detection system. It uses facial landmark models, such as eye corners, nose tip, and/or face appearance models.
For the input image, the following extensions are supported: JPG
, JPEG
, and PNG
. The maximum image size is 1920x1080.
[
{
"probability": 0.9999996423721313,
"rectangle": {
"left": 89.36750197336077,
"top": 57.552113214756055,
"right": 293.84930589000385,
"bottom": 350.8455882774045
},
"direction": {
"yaw": 0.9754817470297894,
"pitch": -1.840920759566032,
"roll": 0.9923179751837665
}
}
]
The output is a list of dictionaries that include the coordinates of all the faces and the probabilities of detection.
The location of the detected face is presented as the top-left and bottom-right pixel coordinates of the frame.
In the corresponding dictionary of a face, there are values for yaw, pitch, and roll angles for each detected face.
While analyzing your content, you may encounter the errors listed below:
In case of any errors, please send the details to this email: info@smartclick.ai.
We strive to deliver the highest quality of our services to help companies streamline their business processes, improve operational efficiency and maximize profits. To help companies respond to a fast-changing business environment, we offer fully customizable solutions that are more efficient and adaptable to specific business needs and budgets.
Have some questions about APIs?
Contact us:
📩 Email: info@smartclick.ai
Instagram
Facebook
LinkedIn