by default, depth and color images are automatically registered, so you
don't have to do anything special. It can be disabled using
dynamic_reconfigure (see Section 5, setting "depth_registration",
On 05/07/2011 11:42 AM, Boris Lau wrote:
> what is the best way to obtain RGBD images for use in openCV, containing
> color and depth values, both aligned in the color camera frame?
> I'd like to avoid doing this alignment myself, since it's already
> somewhere in the openni_kinect code. But Taking an XYZRGB point cloud
> and transforming it into a depth image sounds like a detour...
> Or should I do this without openni_kinect at all?
> Thanks for any pointers