[ros-users] Working object detection and recognition code

Radu Bogdan Rusu rusu at willowgarage.com
Thu Feb 17 23:52:50 UTC 2011


The code in tod_* is currently undergoing major changes until it will be ready for release. We're doing our best to make 
sure that everything fits nicely together in our REIN (Recognition Infrastructure) framework, but we don't have a 
concrete deadline for when this will be ready.

Bottom line is: there are quite a few novel methods that we worked on in parallel, some using geometry, some using 
texture, that will come together in a nice way in the end. :)

Cheers,
Radu.
--
http://pointclouds.org

On 02/17/2011 03:20 PM, dlmypr wrote:
>
> Hi all,
>
> I use tod_training and tod_detecting for object recognition. It works pretty
> well but the output of the program is just an image showing the estimated
> pose and the name of the recognized object. How can I get this output of
> estimated pose like as a matrix but not just the image?
>
> Also does any one knows any other working! object detection or recognition
> code?
>
> Thanks..



More information about the ros-users mailing list