General Object Recognition based on MobileNet-SSD Model

Execute the following steps to recognize general objects based on MobileNet-SSD Model:

Step:

To run C++ application:

To run the demo code for general object recognition based on MobileNet-SSD on VisionFive 2 Debian, execute the following command under any path under user:

example_3rd_dnn_mobilenet --device=4
Tip: Parameter Information:
  • The parameter device is to set the video device number for capturing the video stream. Normally, /dev/video1 represents the camera connected to MIPI CSI, and /dev/video4 represents the USB camera.
  • Other parameters, such as mode path and thread value settings, are set in the source code, and configuration is not allowed.

Result:

  • HDMI monitor displays the real-time video stream from a webcam;
  • HDMI monitor displays real-time drawings of bounding boxes, labeling the objects recognized by the model, and displaying their names;
  • The inference time per frame is displayed on the upper left corner, which translates to an approximate inference frame rate of 2.5-3 fps;
Figure 1. General Object Recognition based on MobileNet-SSD Model