Is d430 pointcloud output in ros affected by resolution of depth image?
Using realsense-ros package to launch a d430 camera with enable_pointcloud:=true to obtain a point cloud.
My question is if I change the depth_frame size from 640 x 480 to 1280 x 720, will it result in better pointcloud?
Thank you
-
Hi @Db1011 enable_pointcloud is hardly ever used now. The modern way to set up a point cloud in the RealSense ROS wrapper is to include the term filters:=pointcloud in your launch statement.
https://github.com/IntelRealSense/realsense-ros#rgbd-point-cloud
The optimal depth accuracy resolution for the D430 / D435 cameras is 848x480. 1280x720 is the optimal depth resolution on the D415.
You could edit the launch file resolution but most people do not as far as I know, suggesting that there is not an advantage to be gained from doing so.
-
Hi MartyX Grover,
Thank you for information. Default resolution for depth image is set to 640 x 480. I would be trying to change the launch file resolution to 848 x 480 for depth map by using the args tag in the roslaunch.I was using filters:=pointcloud before, but then switched to enable_pointcloud as it felt more indicative and intuitive to me.
I am curious is there a difference between the two ways of getting pointcloud streams?
Thank you
-
enable_pointcloud was deprecated as far back as the previous ROS wrapper realsense_camera for the original generation of RealSense cameras. The documentation says simply that it was set to false by default because of "performance issues" and that using filters:=pointcloud is recommended instead.
Using a deprecated function is not necessarily wrong. It just means that the function may be removed at some future date. Due to the amount of time that has gone by since its deprecation though, it is more likely that the enable_pointcloud command will be left in but not be recommended for use.
-
Thank you MartyG,
Lastly, I have been using this documentation--> https://github.com/IntelRealSense/librealsense/tree/master/doc
Can you please point me to other documentation I should refer?This will help me figure out answers to questions like 'will spatial filter be applied for pointcloud stream/filter?'
-
If you are writing your own scripts in the RealSense SDK, an excellent resource is an unofficial documentation site that draws information directly from the official docs and arranges it into a user-friendly format that is searchable with menus at the top of the page.
https://unanancyowen.github.io/librealsense2_apireference/annotated.html
Also, if you are writing your own script then assume that - unlike the RealSense Viewer - there is no post-processing active by default and you need to deliberately program support for it into your application yourself.
Intel's main post-processing guide documents are in the links below:
https://github.com/IntelRealSense/librealsense/blob/master/doc/post-processing-filters.md
https://dev.intelrealsense.com/docs/depth-post-processing
https://github.com/IntelRealSense/librealsense/tree/master/examples/post-processing
The White Papers archive has a wide range of RealSense papers on various subjects:
https://dev.intelrealsense.com/docs/whitepapers
Intel's official RealSense blog is also a very good source for subject articles:
https://www.intelrealsense.com/blog/
Recordings of past seminar events are available on the Webinars and Events page:
https://www.intelrealsense.com/webinars-and-events/
And there are YouTube videos on the official RealSense YouTube channel:
-
The best source of documentation for the RealSense ROS wrapper is its main page. Please scroll down through its various sections.
https://github.com/IntelRealSense/realsense-ros
You can also ask ROS questions at the official RealSense ROS GitHub forum by visiting the link below and clicking on the New Issue button.
https://github.com/IntelRealSense/realsense-ros/issues
You can do the same for non ROS questions at the main librealsense GitHub:
https://github.com/IntelRealSense/librealsense/issues/
If you have a specific question about librealsense or the ROS wrapper, a good way to research existing answers is to do a web search that starts with the search words librealsense or with librealsense ros
Please sign in to leave a comment.
Comments
7 comments