This Raspberry Pi project is like something straight out of the Predator movie universe. It was created and shared by YouTube channel Engineering After Hours—begging the question, is this what engineers do in their off time? In this case, apparently so.
The project uses a Nerf blaster as housing for a Raspberry Pi face tracking project. The whole unit can be mounted to his shoulder, just like the Predator's shoulder cannon from the 1987 movie. It uses image recognition software to locate and track a target.
The software is set up using Tensorflow on a Raspberry Pi 4 model B. A Google Coral Edge TPU was used to help accelerate the image recognition process. Whenever Tensorflow detects a person, it triggers servo mechanisms to move the camera until it has centered on that person.
To manipulate the servo motors, a Pimoroni Pan-Tilt HAT is used. This makes it easier to control the blaster camera so it can face any direction. The unit is completely portable, relying on batteries for power.
If you enjoyed this project, you should check out the official YouTube channel: Engineering After Hours. There you can find a complete demo of this blaster face tracker and more cool projects to explore.