Our Hi3519-based face detection camera can perform many deep learning tasks within the camera, including detecting each face in the scene, estimating the age, gender, and pose of each face, keeping track of each person, and transmitting the best pose of each person to the server. This way, the entire frame does not need to be sent to the server.
This can significantly reduce the bandwidth requirement of the network and the computational requirement of the server. The server can perform face recognition directly by comparing the input face to those in its face database.
4.1 x 4.1 x 12.5 cm
If your next project requires detection, recognition and tracking, talk to us.
We will work with you in the integration process, and you will learn why our
technology backed by deep-learning is amongst the best in the world.