The working principle of somatosensory game

The principle of somatosensory game: Using high-tech video motion capture technology, the player's body movements or gestures are analyzed through camera data to realize direct human-computer interaction.

Somatosensory game, a common family game, is introduced: Through the link between the game console and the TV, a game console allows players to complete the game through whole body coordination, which is no longer a game controlled only by fingers, thus achieving a more perfect game experience. The introduction of the amusement park somatosensory game, through the high-tech projector installed in the ceiling and other places, players can use the high-tech video motion capture technology to read their actions in real time. Players can control the game simply and directly through physical activities without using a keyboard, joystick or mouse. The interactive system can analyze the actions of many people at the same time, and realize the real effect of multi-person interaction and PK in the same field. (M3 Paradise (wall), M3 Paradise (ground)) also has desktop somatosensory games, but the difference is that the actions are read by the video motion capture system above the desktop screen. (hockey, table tennis)

Somatosensory games are usually divided into two parts: host and control terminal. The host is responsible for running and processing signals; The control terminal is responsible for collecting dynamic signals and transmitting them to the host. There are also technical differences in the principle and implementation of the control terminal. Take Microsoft's somatosensory game XBOX 360 as an example. Its control terminal is called Kinect, which is essentially a 3D somatosensory camera, including an infrared emitter and a camera. The infrared emitter emits "laser" to cover the whole visual range of Kinect, and the camera receives the reflected light to identify the player.

Then, Kinect will look for a moving object that may be a human body in the image, that is, distinguish the human body from the background environment by extracting useful signals from the noise.

Then, each pixel of the player's image is sent to the machine learning system to identify which body part the specific pixel belongs to.

Finally, using the output results of the previous stage, a skeleton system is generated according to the tracked 20 joint points. In this way, human movements are captured.

After getting rid of the shackles of the handle control, not only the hand movements, but also the swing of the head, the rotation of the body and the jump of the steps will be displayed on the screen truthfully, thus realizing a more realistic and intuitive game experience. On the other hand, the game machine has also changed from the criticized sit-in button mode to the whole body exercise and intelligent fitness mode, which is naturally very popular.