Docs on camera matrices, flow, depth updated

[mmbannert]

Hi,

I have improved the README.md of the MOVI demos and removed a couple of mistakes. The README now covers some essential information that will be of interest to many users.

1. As Qwlouse pointed out here, the depth values do not require any rescaling by data_ranges. That discussion further illustrates how the optical flow data can be accessed: They are stored as single-precision floats in the G and R channels of the PNG files. Both depth and optical flow are stored as single-precision floats and not as uint16.
2. The documentation now states how camera quaternions and positions are combined to obtain (possibly more user-friendly) extrinsic camera matrices and how they are correctly interpreted: i.e., that it describes a camera-to-world transformation and where each positive direction in the world coordinate system points. The description is loosely analogous to the extensive docs of this this dataset.
3. The docs now briefly describe the availability of the intrinsic camera matrix and how it can be converted to pixel units.

I hope this helps!

Cheers,
Michael

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[Qwlouse]

Thanks a lot!