Can I directly get the registered point cloud image

Question

Hi， I am a college student from China. I want to Use Azure-Kinect to build a 3D model of an object so I bought 2 Azure-Kinect. One is placed in the middle and the other is placed on the side. How can I obtain a point cloud synthesized by two point clouds without using an algorithm like ICP registration? In other words, does this device support directly output of the synthesized point cloud? If so, please tell me how to achieve it.

Answer 1

@fisher
Here is the response I received from Product Team on your queries.

Q: How can I obtain a point cloud synthesized by two point clouds without using an algorithm like ICP registration?

Ans: You will have to register the cameras with an external target.
See step 3 in this example code https://github.com/not-the-programmer/Azure-Kinect-Sensor-SDK/tree/develop/examples/calibration_registration
You can skip the intrinsic calibration process as the Kinect devices are already calibrated for intrinsics. You can obtain the calib.yaml using k4a_device_get_calibration() function.

Q: In other words, does this device support directly output of the synthesized point cloud?

Ans: No

Do let us know if you have any further queries.

---------------------------------------------------------------

If an answer is helpful, please "Accept answer" or "Up-Vote" for the same which might be beneficial to other community members reading this thread.