[ros-users] vslam_system
Vaibhav Ghadiok
vaibhavghadiok at gmail.com
Sat Jul 3 23:34:41 UTC 2010
These are training dataset learned from images in the holidays dataset. This
should get you more info:
http://cvlab.epfl.ch/~calonder/
I have not had the time to figure this out.
I was able to find the tree file from the willow garage website, but not the
weights one. I had posted a message but got no response.
Do let me know if you figure it out. I believe one would have to download
the dataset and use a training program to form the basis vectors to figurer
out image signatures.
On Sat, Jul 3, 2010 at 5:18 PM, hudvin <hudvin at gmail.com> wrote:
>
> If I remember right, this is because it is not being able to find the
> place_recognizer training files holidays.tree and holidays weight.
>
> place_recognizer_("/u/mihelich/vocab/holidays.tree",
> 17 "/u/mihelich/vocab/holidays.weights"),
>
> Looks like local paths. Is this images for test purposes or something else?
>
> >>Your cam params certainly don't seem right. I guess you put something
> just
> to make it run.
> Yes, just for test.
> --
> View this message in context:
> http://ros-users.122217.n3.nabble.com/Example-of-SLAM-using-mono-or-stere-camera-tp920634p941447.html
> Sent from the ROS-Users mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> ros-users mailing list
> ros-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ros-users
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
--
http://guidance.aero.iisc.ernet.in/vaibhav
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20100703/34ed006b/attachment-0003.html>
More information about the ros-users
mailing list