Intrinsic Camera Calibrator: glog flag conflict error (logtostderr defined twice)

[Ablazesphere]

I'm following the intrinsic camera calibration tutorial (https://autowarefoundation.github.io/autoware-documentation/main/how-to-guides/integrating-autoware/creating-vehicle-and-sensor-model/calibrating-sensors/intrinsic-camera-calibration/) and getting this error when launching:

ERROR: flag 'logtostderr' was defined more than once (in files './src/logging.cc' and '/home/shuttle_01/Desktop/Reva/autoware.reva/src/universe/external/glog/src/logging.cc').
[ERROR] [camera_calibrator-1]: process has died [pid 101120, exit code 1]

It looks like there's a conflict between the system glog (likely pulled in via libgoogle-glog-dev or Ceres) and Autoware's external glog package. Both are being linked, causing duplicate flag definitions.

Environment:

• Autoware.reva workspace
• Following the official intrinsic camera calibration tutorial
• The ceres_intrinsic_camera_calibrator package depends on libgoogle-glog-dev

Has anyone encountered this before? Any suggestions on how to resolve the glog linking conflict? 🤔

Thanks in advance!

[paulsohn]

As a temporary workaround, it might be helpful to use the standalone version outside from the autoware directory.

[Ablazesphere]

As a temporary workaround, it might be helpful to use the standalone version outside from the autoware directory.

Yes, that resolved the glog conflict issue. However, I'm now encountering a problem with the intrinsic camera calibrator—the visualization overlays aren't appearing on my video feed.

[Ablazesphere]

As a temporary workaround, it might be helpful to use the standalone version outside from the autoware directory.

Yes, that resolved the glog conflict issue. However, I'm now encountering a problem with the intrinsic camera calibrator—the visualization overlays aren't appearing on my video feed.

Everything is working now. I made a mistake in setting up the chess board parameters.

My bad. Thank you!

[paulsohn]

Technically speaking, the glog conflict is not officially 'resolved' (the solution I said is a mere workaround) so I think the issue seems worth keeping open. Though it's up to you to decide whether to reopen this or not.