How can I save the complete frameset in c++ using the Intel RealSense SDK?

[LLits]

I would like to know if it is possible to store the complete frameset using the SDK obtained from the wait_for_frames() function. At the moment I save the individual raw images, but it would be much more convenient to save the whole frameset. Is there a option to do this or do you have other suggestions. I would like to save them such that I can perform color alignment and filtering later on. Thank you in advance!

[MartyG · Accepted solution]

An option for storing all the frames may be to use the SDK's record function to record a type of file called a 'bag' (also known as rosbag), which stores all the frames in a stream and can be played back from the file afterwards.

Details of the SDK's record and playback can be found on the documentation page linked to below.

https://github.com/IntelRealSense/librealsense/blob/master/src/media/readme.md

Dorodnic the RealSense SDK Manager has said on this subject: "Best way to save RealSense data without losing information would be the ROS-bag format. Our built-in recorder will collect everything you need, and it is fairly well optimized ... you can then extract individual frames / models as a post-processing step".

A bag can also be loaded into the RealSense Viewer program instead of using a live camera stream, so that you can do further work with that data set.

https://github.com/IntelRealSense/librealsense/blob/master/doc/sample-data.md

Bag files can be quite large, going into the gigabytes, so it is recommended that you create a folder for the files on your computer and set a custom save path to send the files to that folder.