TROS Body Detection

Application Scenarios

RDK Studio helps beginners get started quickly and enables an efficient workflow for human body keypoint detection: The human body keypoint detection algorithm is primarily used to capture skeletal keypoints of the human body, suitable for multi-person pose recognition tasks in real-time application scenarios.

Preparations

Supports connecting both USB and MIPI cameras. This section uses a USB camera as an example. The connection method for a USB camera is as follows:

Running Process

1. Click Node-RED under the TROS Body Detection example.

   

2. Enter the example application flow interface.

   Tip

   Click the icon in the top right corner of RDK Studio to quickly open the example in a browser!

   

3. Select the type of camera connected, click the corresponding Start(USB Cam) command, and wait for about 10 seconds for the visualization window to open automatically.

   

4. Performance Result: Click the debug icon on the right to position the sidebar to the debug window. You can view the performance information output results in the debug window.

   

5. Output Statistics: Can output statistical results of detected human body keypoints.

   

6. Click to execute the Stop command to turn off the camera.

   

   Note

   If you have modified nodes, flows, etc., you need to click the button in the upper right corner for the changes to take effect!

7. Click the × icon in the upper right corner，Select "Close APP" to exit the Node-RED application.

   

More Features

Visualization Page

1. Click to execute the Visualization Interface command to automatically open TogetherROS Web Display.

   

2. Click Web Display to enter the visualization page for real-time target detection.

3. Click the × in the upper right corner of the visualization page to exit.

Learn More

Click to execute the Learn More command to open the online documentation to find more information about the examples.