Stereo visual odometry

Hi everybody. It’s been quite a long time without writing anything. I’ve been incredibly busy lately with the projects reaching the integration stages, preparing documentation, etc.

What I can offer today is a video that summarises what we’ve been doing in Vicomtech about stereo odometry.

Actually this video shows our algorithm for egomotion estimation applied on one of the sequences publicly available by the guys in the Karlsruhe University

Remarkably, we are strongly focused on the exploitation of OpenCV and PCL primitives.

In the next months we will be publishing our own databases and stereo videos, and soon, our firsts results in SLAM.

Hope you like it!

This entry was posted in Computer vision and tagged , , , , , . Bookmark the permalink.

6 Responses to Stereo visual odometry

  1. leon says:

    Does it now have a release of OpenCV?

  2. faizan says:

    What kind of egomotion estimation algorithm are you applying ?? Are you just trying to find Essential matrix and then finding R & T matrix from there. Any papers would be helpful.

    • Hi!
      We are triangulating 3D points from epipolar correspondences and then applying resectioning. We are preparing the paper that describes this method. I hope we can publish it soon.
      Best regards

  3. shashank Bhushan says:

    Hii sir, i am a student final year undergrad student and I want to implement Stereo slam as a part of my project.It will be of great help if you suggest some paper and relevant code for my project .May I know which filter have you implemented in case of this Stero SLam and Which method you used for loop closures.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s