Hello Issac Chan,
I hope below information may help with your initial query!
References:
Body Tracking SDK uses a body tracker object to process Azure Kinect DK captures and generates body tracking results. It also maintains global status of the tracker, processing queues and the output queue. There are three steps in using the body tracker:
- Create a tracker
- Capture depth and IR images using Azure Kinect DK
- Enqueue the capture and pop the results.
Enqueue the capture and pop the results
You can use the k4abt_tracker_enqueue_capture()
function to add a new capture to the input queue and the k4abt_tracker_pop_result() function to pop a result from the output queue.
The k4a_device_get_capture()
function can be used to capture a depth frame. The K4A_DEPTH_MODE_NFOV_UNBINNED
or K4A_DEPTH_MODE_WFOV_2X2BINNED
modes are recommended for best performance and accuracy.