View my account

"Using Pre-Recorded Data for Box Measurement and Multi-Camera Calibration in Python"

Follow
Araj7322

Hello,

I am working with the Box Measurement and Multi-Camera Calibration example provided in the Intel RealSense Python code. I would like to know if it is possible to measure the dimensions of an object using this example without having access to a RealSense camera.

Specifically, I am exploring whether pre-recorded data (e.g., .bag files) can be used instead of live camera feeds to perform the measurements.

Here are my key questions:

  1. Can the example code be modified to work with .bag files as input instead of live camera streams?
  2. If yes, what changes would be required in the code to enable this?

I am using the following setup:

  • Python Version: Python 3.10.12
  • Operating System: Ubuntu 22.04.5 LTS

Any guidance, suggestions, or references to similar implementations would be greatly appreciated.

Thank you in advance for your help!

0

Comments

5 comments

Date Votes
  • MartyG

    Hi Araj7322  You can use a pre-recorded bag file as the data source in a script instead of a live camera simply by inserting an enable_device_from_file() config instruction directly before the pipeline start line of the script.  This principle is demonstrated in the RealSense SDK's Python example script read_bag_example.py

    https://github.com/IntelRealSense/librealsense/blob/master/wrappers/python/examples/read_bag_example.py#L42

    0
    Comment actions Permalink
  • Araj7322

    Thank you, MartyG. Can we calculate the 3D dimensions of an object using the Box Dimensioner Multi-Cam code in Python with a pre-recorded .bag file? If so, how can we implement that? Also, since only limited .bag pre-recorded files are available, are there any additional .bag files available that can be used for this task?

    0
    Comment actions Permalink
  • MartyG
    • Edited

    I cannot recall a previous case where bag files have been used instead of a live camera with the box_dimensioner_multicam project.  It may still work though if you try adding a bag import instruction to the project's code.

     

    Some official sample bag files for testing can be found at the link below.

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

     

    I would expect though that those bags would not be useful for use with box_dimensioner_multicam though as they are not focused on a box object.

    0
    Comment actions Permalink
  • Araj7322

    Thank you MartyG for your response.

    0
    Comment actions Permalink
  • Neighsanlo
    • Edited

    This is useful for me.

    0
    Comment actions Permalink

Please sign in to leave a comment.

Didn't find what you were looking for?

New post