add imagedata process pipeline for gnssfgo

• input

  • cv::Mat

• output

  • Pose / factor / interaction with gnssfgo / visualization

• build image data ptr and hand the ptr to odom for processing

  • Recover the pose by PnP or other algorithm, filter the match ing findEssentialMat RecoverPose and solve PnP
  • Manage the feature using FeatureManager
  • Record following data
    • # of feature of this frame
    • # of trakced feature of this frame
    • # of inlier matches of this frame

• display image with feature using built ptr

  • yellow represent new feature
  • red represent matched feature
  • Blue represent matched with existed landmark feature

• add a demo (optional with ROS2)