Skip to content

Synchronization of image and IMU and output a dataset viable for deep learning

Notifications You must be signed in to change notification settings

BlanchonMarc/Sync_Pola_IMU_ROS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

Collect data from Pola + IMU

Collect data from Pola + IMU


Polarcam

  • Connect the Polarcam G1 to the computer. Do not forget it's POE.
  • Test the Camera stream.

Initialize the Camera node:

roslaunch pylon_camera pylon_camera_node_G1.launch

Check the image_raw stream:

rqt_image_view


IMU

  • Connect the IMU to the computer. (You can also follow instructions here if not installed)
  • Test the IMU stream.

Initialize the IMU:

roslaunch euler_imu imu_gx3-35_init.launch --screen

Start the IMU:

roslaunch euler_imu imu_gx3-35_full_start.launch --screen

Check the pose stream of the IMU:

rostopic echo /imu_3dm_node/imu/pose


Record

To record a bag containing all the current information, just create it through:

rosbag record -a -o [Name_of_Your_Bag]


Extract the data from the recorded bag

You first need to have a bag recorded containing an IMU and a Camera stream. Refers to previous parts if you need to record bag. When you have the bag you just have to specify the bag, the topics and the output folder.

Exemple:

./ros/kinetic/catkin_ws/src/Kalibr/aslam_offline_calibration/kalibr/python/kalibr_bagextractor --bag PolarIMU.bag --image-topics /pylon_camera_node/image_raw --imu-topics /imu_3dm_node/imu/data --output OutputIMU2/

It will result of a folder created in the folder you are and inside you will have the images and the csv file:

-> /home/mscv/
    |-> PolarIMU.bag
    |-> OutputIMU2/
        |-> cam0/
            |-> 143763883.png
            |-> 143763882.png
            |-> ...
        |-> imu0.csv

Organize to build the dataset

Execute the organizer giving the folder of image, the csv file and the iutput folder for the separated csv.

Exemple:

python organize.py --input OutputIMU2/cam0 --csv OutputIMU2/imu0.csv --output OutputIMU2/csv

It will output synchronized imu and the synchronization. At the end, you'll have image and correspoinding imu with the same filename.

-> /home/mscv/
    |-> PolarIMU.bag
    |-> OutputIMU2/
        |-> cam0/
            |-> 143763883.png
            |-> 143763882.png
            |-> ...
        |-> imu0.csv
        |-> csv/
            |-> 143763883.csv
            |-> 143763882.csv
            |-> ...

About

Synchronization of image and IMU and output a dataset viable for deep learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages