Configuring the Person location tracker module

You can configure the Person location tracker module on the settings panel of the Person location tracker object that is created on the basis of the Camera object on the Hardware tab of the System settings window.

To configure the Person location tracker software module, do the following:

1. Go to the settings panel of the Person location tracker object.
   
2. Set the Show objects on image checkbox if you want the detected object to be outlined when you view live video.

3. From the Device drop-down list, select the device on which the neural network operates. Auto—device is selected automatically: the NVIDIA GPU gets the highest priority, followed by the Intel GPU, and then the CPU.
   

   Attention!

   • We recommend using the GPU.
   • It can take several minutes to launch the algorithm on the NVIDIA GPU after you apply the settings. You can use caching to speed up future launches (see Configuring the speedup of neural analytics launch on GPU).
4. From the Neural network for pose estimation drop-down list, select one of four detector modes:
   1. Person pose estimation—The standard neural network is used. The default value.
   2. Person pose estimation (Nano)—A small neural network is used.
   3. Person pose estimation (Middle)—A medium neural network is used.
   4. Person pose estimation (Large)—A large neural network is used.
5. If you want to use a custom neural network, click the  button in the Tracking model field and specify its file in the standard Windows Explorer window. If you leave the field blank, the default networks are used. They are selected automatically depending on the selected device.

6. In the Frames processed per second [0.016, 100] field, set the number of frames per second that the detector processes.

   Attention!

   For the scenes with static people, the FPS value must be at least 2. For the scenes with moving people—at least 4.

   The larger this value, the higher the accuracy of pose detection, but the greater the CPU load. With FPS=1, the accuracy is no less than 70%.

   This parameter varies depending on the object's speed of movement. To solve typical tasks, the FPS value from 3 to 20 is sufficient. Examples:

   • Pose detection for moderately moving objects (without sudden movements)—3.
   • Pose detection for moving objects—12
7. In the Track retention time (s) field, specify the time in seconds in the range from 0.3 to 1000 after which the object track is considered lost. This parameter is useful in situations when one object in the frame temporarily overlaps another.
   

   Note

   If an object (track) is close to the frame border, then approximately half the time specified in the Track retention time (s) field must elapse from the moment the object disappears from the frame until its track is deleted.

8. Click the Settings button. As a result, the Detection settings window opens.
   

   1. Click the Stop video button (1) to capture a frame.

   2. Click the Area of interest button (2) and specify the area in which the person location tracker operates (3). The specified area is outlined with a red frame.
      

      Note

      To delete the selected area, click the button.

   3. Click the OK button (4) to save the detector settings.

9. Click the Apply button on the settings panel of the Person location tracker object.

10. Configure the Pose detection (see Configuring the Pose detection), the Shooter detection (see Configuring the Shooter detection), or the Handrail grip detection (see Configuring the Handrail grip detection).

Configuring the Person location tracker module is complete.