I want to ask about the usability of IMU data. Can I directly feed the imu data into a Visual-Inertial System? Do I have to do the transformation on IMU data before I feed the data? Addition: I visualize the trajectory of sponza. The gravity is in negative z direction. However, in my recorded rosbag, the negative gravity is added to y component of accelerometer measurement. Is it reasonable?
Besides, I would also like to ask why the camera extrinsic (rotation matrix) in calibration file always in [[1 0 0] [0 -1 0] [0 0 -1]], they have -1 in the y and z component?
Hi,
I want to ask about the usability of IMU data. Can I directly feed the imu data into a Visual-Inertial System? Do I have to do the transformation on IMU data before I feed the data?
Addition: I visualize the trajectory of sponza. The gravity is in negative z direction. However, in my recorded rosbag, the negative gravity is added to y component of accelerometer measurement. Is it reasonable?
Besides, I would also like to ask why the camera extrinsic (rotation matrix) in calibration file always in [[1 0 0] [0 -1 0] [0 0 -1]], they have -1 in the y and z component?
With best regards, Yuxuan