Replies: 3 comments 1 reply
-
Hi @epanov RealSense cameras use Y for the vertical axis and Z for the forward axis. The ROS standard is X-forward, Y-left, Z-up. I note that your Fixed Frame in RViz is set to Ethernet0/gpe. When visualizing a RealSense pointcloud in RViz, Fixed Frame will more commonly be set to camera_link If using camera_link is not possible for you and you need to change the axes, you could explore the possibility of using a static transform or alternatively, use the advice in the link below to reconfigure the position / orientation of axes in RViz. https://answers.ros.org/question/339109/how-to-change-the-cameras-view-in-rviz-in-the-begin/ |
Beta Was this translation helpful? Give feedback.
-
After coming back to my work and restarting the nodes it appears that things work now. Will close thread. Apologies for the false alarm. |
Beta Was this translation helpful? Give feedback.
-
It's no trouble at all. I'm pleased to hear that it's working for you. Thanks very much for the update! |
Beta Was this translation helpful? Give feedback.
-
In the picture below the raw depth and the aligned to rgb depth is shown on top of a point cloud from an rs455 using the depth_aligned topic.
The picture below shows the
aligned_depth_to_color
topic on top of the rs455 point cloud. Note the stretching of this point cloud.Note how the depth, x axis in workspace, z axis in realsense image, is accurate. This to me implies that the xdpi and ydpi may be incorrect.
Looking at the ros2 param.s they are set to default since
ros2 param get /camera/hand_realsense .hand_realsense.color.image_raw.tiff.xdpi
andydpi
return -1.For now I will use the depth/image_rect_raw topic with intensity, but it would be ideal to have an accurate rgbd image.
EDIT:
Added context for second image.
Added more accurate axis names.
Beta Was this translation helpful? Give feedback.
All reactions