Hi @ZH Chang Greetings! Welcome to Microsoft Q&A forum. Thank you for posting this question here.
The accuracy of the depth camera on Azure Kinect which is responsible for body tracking depends on additional factors in addition to the distance such as lighting conditions, the number of people in the scene and the complexity of their movements among other things.
Unfortunately, there is no specific recommended setting that is documented to give a better capture from Azure Kinect. It has to be drawn to a conclusion just like you did, based on observations and practical experience. I am glad to know that you have arrived at a conclusion that 0.5*(l_min+l_max) seems to work best.
Unfortunately, there is no specific recommended setting that is documented to give a better capture from Azure Kinect. It has to be drawn to a conclusion just like you did, based on observations and practical experience. I am glad to know that you have arrived at a conclusion that 0.5*(l_min+l_max) seems to work best.
Please find the below image that captures the Depth camera supported operating modes and the operating range for each setting.
Please find the below image that captures the Depth camera supported operating modes and the operating range for each setting.
This might help you test different operating modes and figure out what might be a better range given the mode of operation.
Hope this helps.
If the response helped, please do click Accept Answer and Yes for the answer provided. Doing so would help other community members with similar issue identify the solution. I highly appreciate your contribution to the community.