Issue with using negative disparity in stereo_image_proc
When I set the min_disparity value to a negative number in stereo_image_proc, the resulting point cloud has points with negative z values, as seen in RVIZ. This seems like a bug. Has anyone else encountered this problem?