Class yolomodel::tensorflow::TPUInterpreter: This class correctly interfaces with the TPU, loads models, and runs inference, but any attempt to parse the output received from the result of the inference is garbage.
Member ZEDCam::SetPositionalPose (const double dX, const double dY, const double dZ, const double dXO, const double dYO, const double dZO) override: The ZEDSDK currently cannot handle resetting the positional pose with large translational (dX, dY, dZ) values without breaking positional tracking. This is because the values are floats and not doubles. To fix this I (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om), have decided to just handle the translation offsets internally. So when SetPositionalPose() is called is assigns the dX, dY, and dZ values to private member variables of this class, then the offsets are added the the pose in the PooledLinearCode() under the pose requests section. If StereoLabs fixes this in the future, I will go back to using the sl::Translation to reset positional tracking.