atom
atom copied to clipboard
ATOM only capable of handling plumb_bob distortion model
As I was trying to collect data (#935), the rgb_labeler
crashes, as the distortion model of the T265 camera (equidistant
) is not compatible with the assumptions used by this script.
As we have not tried to do anything with this type of camera, I think more errors along the pipeline are to be expected.
What would be the best way to add this to the dataset?
Should I add a new parameter to the camera_info, with the distortion model?
Then all calculations can be done according to the model?
Tagging @miguelriemoliveira and @Kazadhum for visibility.
My bad, the dataset already has the full camera info. However, some calculation might be at stake. I will keep an eye on this.
Hi @manuelgitgomes,
my opinion is that for now if you can skip this better.
Even if the dataset for now does not contain the fisheye images.
Hello. I have done a small change and this does not crash anymore on data collection. Let's see in calibration.