3D camera technology represented by ToF and structured light is the best match for the needs of face recognition technology. First of all, the 3D camera uses infrared rays as the emitted light, which can solve the problem of the environmental lighting effect of visible light. When the traditional 2D recognition technology changes in ambient light, the recognition effect will drop sharply, which cannot meet the needs of the actual system. For example, it may not be recognized correctly when it encounters side light when taking pictures.
ToF or structured light 3D camera technology collects face image depth information, which can obtain more feature information and greatly improve the recognition accuracy on the basis of traditional face recognition technology. Compared with 2D face recognition system, 3D face recognition can collect depth feature information such as eye canthus distance, nose tip point, nose wing point, distance between two temples, distance from ear to eye, etc.3D face recognition can continue to maintain a very high recognition accuracy when user characteristics change.
Why do AR/VR devices use 3D ToF cameras? ——1. Obtain RBG data and depth data of surrounding environment images, and perform 3D reconstruction; 2. Realize human-computer interaction methods such as gesture recognition and motion capture.
The 3D perception of AR/VR generally adopts 3D ToF camera active perception technology in principle. The front of the device usually includes an infrared transmitter, an infrared sensor (to obtain depth information) and multiple ambient light cameras (to obtain RBG information). Taking 3D ToF camera technology as an example, the infrared transmitter emits infrared rays, which are reflected by the target object and received by the infrared sensor. The phase difference between the transmitted signal and the received signal is used for calculation and conversion to obtain distance/depth of field data. The gesture recognition function provided by the 3D ToF camera will become the core interaction method in the AR/VR field in the future. There are rich human-computer interaction content in the field of AR applications, and this interaction is very complex, and the 3D tof camera technology can be fully applied.