face recognition manufacturer < / a > on face attribute recognition technology and algorithm

"Face attribute recognition" is a technology to recognize the attribute values of human face, such as gender, age, posture, expression and so on

The input of general face attribute recognition algorithm is "a face image" and "coordinates of key points of facial features", and the output is the corresponding attribute value of face. Face attribute recognition algorithms generally align the face according to the coordinates of key points of facial features (adjust the face to the predetermined size and shape after rotation, scaling, deduction and other operations), and then conduct attribute analysis

Conventional face attribute recognition algorithms recognize each person's face attributes as an independent process, that is, face attribute recognition is only a general term for a class of algorithms. Gender recognition, age estimation, pose estimation and expression recognition are all independent algorithms. However, some face attribute recognition based on deep learning also has the ability of an algorithm to input attribute values such as gender, age and posture at the same time