Kinect calibration - why input depth in meters

[Tiến Đạt Ngô]

Dear all,

The documentation about Kinect calibration at http://nicolas.burrus.name/index.php/Research/KinectRgbDemoV4?from=Research.KinectRgbDemo says the input to the calibration is the depth images in METERS. So it assumes that the depth computed by Kinect is already accurate? So what is the point of depth calibration and how are the depth images used in calibration because we want to get better conversion from disparity to depth by estimating c0 and c1 parameters of the conversion.

Thank you for your time answering my question.

[Eric Liu]

Hi: I saw your comment, and looked at the code, and I also have the same question that why the depth.raw image data is depth in METERS, and not using the raw disparity data. Have you have found the answers to this yet?

Thanks!