Hi Ivan,<br>You want the posest package, actually - visodo no longer exists.<br>Victor has been working on pose estimation from monocular images, and you can test it out for yourself with the run_mono executable in vslam_system to run on stored image files (or the mono_vslam_node to run on data from ROS topics). However, few motion cases have been implemented; it appears to work well on pure translation, though Victor would know more about it than me.<br>
Thanks,<br>Helen<br><br><div class="gmail_quote">On Fri, Aug 13, 2010 at 2:43 PM, Ivan Dryanovski <span dir="ltr"><<a href="mailto:ivan.dryanovski@gmail.com" target="_blank">ivan.dryanovski@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Hi,<br>
<br>
I am interested in using the vision stack for getting 3D/6DoF pose<br>
estimates from a mono camera. From what I found, visodo is the package<br>
for that. Is the package currently at a (reasonably) usable stage?<br>
<br>
Miguel, Victor, or anyone else who has tried it - what is your<br>
progress and impressions?<br>
<br>
Thank you,<br>
<br>
Ivan Dryanovski<br>
<br>
On Wed, Jun 9, 2010 at 11:57 AM, Abe Bachrach <<a href="mailto:abachrach@gmail.com" target="_blank">abachrach@gmail.com</a>> wrote:<br>
> awesome!<br>
> thanks,<br>
> -=Abe<br>
><br>
> On Wed, Jun 9, 2010 at 11:40 AM, Victor Eruhimov <<a href="mailto:relrotciv@googlemail.com" target="_blank">relrotciv@googlemail.com</a>><br>
> wrote:<br>
>><br>
>> Hi, yes, this is on todo list. For now, you can download the files from<br>
>> (thanks to Patrick!)<br>
>> <a href="http://pr.willowgarage.com/data/calonder_descriptor/current.rtc" target="_blank">http://pr.willowgarage.com/data/calonder_descriptor/current.rtc</a><br>
>> <a href="http://pr.willowgarage.com/data/vocabulary_tree/holidays.tree" target="_blank">http://pr.willowgarage.com/data/vocabulary_tree/holidays.tree</a><br>
>> <a href="http://pr.willowgarage.com/data/vocabulary_tree/holidays.weights" target="_blank">http://pr.willowgarage.com/data/vocabulary_tree/holidays.weights</a><br>
>><br>
>> Best Regards, Victor<br>
>><br>
>><br>
>><br>
>> On Wed, Jun 9, 2010 at 6:46 PM, Abe Bachrach <<a href="mailto:abachrach@gmail.com" target="_blank">abachrach@gmail.com</a>> wrote:<br>
>>><br>
>>> Hey Kurt,<br>
>>> this is probably on your guys' todo list for documentation/release, but<br>
>>> currently, there are a couple issues with the external files required by the<br>
>>> calonder features, and the place recognition...<br>
>>> calonder features:<br>
>>> the current.rtc file is not automatically downloaded when you build the<br>
>>> vslam_system package. To get it, I had to manually run<br>
>>> # make download_data_current.rtc<br>
>>> in the folder<br>
>>> stacks/visual_feature_detectors/calonder_descriptor/build<br>
>>> Furthermore, the path to the current.rtc file is hardcoded into the files<br>
>>> vision/vslam_system/run_stereo.cpp<br>
>>> vision/frame_common/src/frame.cpp<br>
>>><br>
>>> place recognition:<br>
>>> the path to the holidays.tree and holidays.weights is hardcoded to<br>
>>> "/u/mihelich/vocab/holidays.tree", "/u/mihelich/vocab/holidays.weights"<br>
>>> in vslam.cpp (and possibly elsewhere)<br>
>>> these files don't seem to have a make rule that allows them to be<br>
>>> downloaded... are they available from somewhere?<br>
>>><br>
>>> finally, the software does not seem to check for the existence of these<br>
>>> files before trying to load them, and does not seem to check whether it was<br>
>>> able to successfully load them.<br>
>>> thanks,<br>
>>> -=Abe<br>
>>><br>
>>><br>
>>> On Wed, Jun 9, 2010 at 2:25 AM, Victor Eruhimov<br>
>>> <<a href="mailto:relrotciv@googlemail.com" target="_blank">relrotciv@googlemail.com</a>> wrote:<br>
>>>><br>
>>>> Hi, yes, this was my change causing errors -- I am in the process of<br>
>>>> integrating mono visual odometry. The compilation problem is fixed in<br>
>>>> r38197. Best Regards, Victor<br>
>>>><br>
>>>> On Jun 9, 2010, at 9:40 AM, Kurt Konolige wrote:<br>
>>>><br>
>>>> > Miguel, the VSLAM stack is experimental at this stage. We'll try to<br>
>>>> > have a fully documented release by the end of the summer. At this point,<br>
>>>> > there are four packages:<br>
>>>> ><br>
>>>> > sba - sparse bundle adjustment<br>
>>>> > posest - pose estimation, 3d and 2d<br>
>>>> > frame_common - front-end for camera parameters, matching<br>
>>>> > vslam_system - system-level integration<br>
>>>> ><br>
>>>> > These should compile correctly. I just tried vslam_system and you're<br>
>>>> > right, it doesn't - I've cc'd Victor who I think made the changes that<br>
>>>> > caused the problem.<br>
>>>> ><br>
>>>> > Cheers --Kurt<br>
>>>> ><br>
>>>> > On 6/8/2010 5:18 AM, Miguel Juliá wrote:<br>
>>>> >><br>
>>>> >> Hi all,<br>
>>>> >><br>
>>>> >> I am trying to use your visual odometry code to test its<br>
>>>> >> implementation<br>
>>>> >> in the vision stack. What is the state of development of this stack?<br>
>>>> >> The<br>
>>>> >> documentation seems to be incomplete and there is no tutorial, just<br>
>>>> >> the<br>
>>>> >> code API. This stack includes visual odometry, place recognition and<br>
>>>> >> toro, doesn't it?<br>
>>>> >><br>
>>>> >> I have downloaded the vision stack and tried to compile it but<br>
>>>> >> vslam_system return some compilation errors. It seems that the posest<br>
>>>> >> package in the repository is different from what vslam_system<br>
>>>> >> expects.<br>
>>>> >><br>
>>>> >> Any advice?<br>
>>>> >><br>
>>>> >> Thanks,<br>
>>>> >><br>
>>>> >> Miguel Juliá.<br>
>>>> >><br>
>>>> >><br>
>>>> >><br>
>>>> >><br>
>>>> >> --<br>
>>>> >> Miguel Juliá Cristóbal<br>
>>>> >> Laboratorio de Automatización Robótica y Visión por Computador.<br>
>>>> >> (ARVC)<br>
>>>> >> Departamento de Ingeniería de Sistemas Industriales.<br>
>>>> >> Universidad Miguel Hernández.<br>
>>>> >> Av. Universidad s/n Edif. Quorum V.<br>
>>>> >> 03202 Elche-Alicante.<br>
>>>> >> Tlf.: 96522-2435<br>
>>>> >><br>
>>>> >><br>
>>>> >><br>
>>>> >> _______________________________________________<br>
>>>> >> ros-users mailing list<br>
>>>> >> <a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
>>>> >> <a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
>>>><br>
>>>> _______________________________________________<br>
>>>> ros-users mailing list<br>
>>>> <a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
>>>> <a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
>>><br>
>>><br>
>>> _______________________________________________<br>
>>> ros-users mailing list<br>
>>> <a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
>>> <a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
>>><br>
>><br>
>><br>
>> _______________________________________________<br>
>> ros-users mailing list<br>
>> <a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
>> <a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
>><br>
><br>
><br>
> _______________________________________________<br>
> ros-users mailing list<br>
> <a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
> <a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
><br>
><br>
_______________________________________________<br>
ros-users mailing list<br>
<a href="mailto:ros-users@code.ros.org" target="_blank">ros-users@code.ros.org</a><br>
<a href="https://code.ros.org/mailman/listinfo/ros-users" target="_blank">https://code.ros.org/mailman/listinfo/ros-users</a><br>
</blockquote></div><br>