Re: [Ros-kinect] Kinect Data Logging

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [Ros-kinect] Kinect Data Logging

Josh Faust
I have a fix (I think) for it completely locking up which I'll try to release tomorrow.  In the mean time, using the "Points" display instead of "Billboards" should help a lot.

Josh

On Thu, Nov 18, 2010 at 2:29 PM, Josh Faust <[hidden email]> wrote:
That change isn't actually going to help much with the Kinect, since there are relatively few invalid points.  I'm going to have to do some profiling to see what's up -- we haven't had a sensor that produces 300k points at 30hz before.

Josh


On Thu, Nov 18, 2010 at 2:12 PM, Radu Bogdan Rusu <[hidden email]> wrote:
Dejan,

Just a bit of time until we make a release. :)

Cheers,
Radu.


On 11/18/2010 02:06 PM, Dejan Pangercic wrote:
> Hi Radu,
> what do we need to do to get the fix for rviz?
> D.
>
> On Thu, Nov 18, 2010 at 11:03 PM, Radu Bogdan Rusu
> <[hidden email]>  wrote:
>> This is a bug we just fixed in RViz.
>>
>> PointCloud2 data can contain NaN numbers, and that was hosing things. RViz is now clever and displays the point clouds
>> without any issues.
>>
>> Alternatively, for a lightweight issue you can do:
>>
>> $ rosrun pcl_tutorials pointcloud_online_viewer input_topic:=/your_topic
>>
>> to visualize PointCloud2 messages. No TF needed! Beautiful! :)
>>
>> Cheers,
>> Radu.
>>
>>
>> On 11/18/2010 01:53 PM, [hidden email] wrote:
>>> My bag files are huge. It usually only takes ~20 seconds before it
>>> reaches the limit. Is there any way to decrease the amount of points
>>> being recorded from the sensor? I know we can do some post processing
>>> filtering with PCL but want to be able to make longer recordings. Is
>>> anyone else having problems with RVIZ? I find that looking at the
>>> point cloud basically seizes up my machine.
>>>
>>> Thanks
>>> Ralph
>>>
>>> On Nov 18, 4:32 pm, Radu Bogdan Rusu<[hidden email]>    wrote:
>>>> If you want to export them from BAGs into standalone PCD files (one PointCloud per file) use:
>>>>
>>>> $ rosrun pcl_ros bag_to_pcd<bagfile>    <topic>    <output_directory>
>>>>
>>>> Cheers,
>>>> Radu.
>>>>
>>>> On 11/18/2010 01:09 PM, binarymillenium wrote:
>>>>
>>>>
>>>>
>>>>> There are these files in the ROS bag format:
>>>>
>>>>> https://code.ros.org/svn/data/trunk/ccny/
>>>>
>>>>> (and more infohttp://www.ros.org/wiki/kinect_node)
>>>>
>>>>> -Lucas
>>>>
>>>>> On Thu, Nov 18, 2010 at 12:48 PM, yair reshef<[hidden email]<mailto:[hidden email]>>    wrote:
>>>>
>>>>>       missed your post *shame*
>>>>>       and so, i second
>>>>
>>>>>       On Thu, Nov 18, 2010 at 5:58 AM, Ashwin<[hidden email]<mailto:[hidden email]>>    wrote:
>>>>
>>>>>           Hi all,
>>>>
>>>>>              I was wondering if it would be a good idea to make logs  of all the
>>>>>           raw data available from Kinect ( rgb and point clouds.) If some has
>>>>>           integrated it with a mobile robot, any physical state information
>>>>>           would be even better.
>>>>>              An immediate benefit of this would that even people who haven't yet
>>>>>           got their hands on the device might still be able to explore its
>>>>>           capabilities. It future it can also help to make the testing phase and
>>>>>           also be a means to benchmarking different approaches in common domain.
>>>>
>>>>>           ~Ashwin
>>>>
>>>>>       --
>>>>>       yair99@gmail
>>>>>       050-6301212
>>>>>       tlv, israel- Hide quoted text -
>>>>
>>>> - Show quoted text -
>> _______________________________________________
>> Ros-kinect mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-kinect
>>
>
>
>
_______________________________________________
Ros-kinect mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-kinect



_______________________________________________
Ros-kinect mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-kinect
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [Ros-kinect] Kinect Data Logging

Kurt Konolige
Yes, you have - the narrow stereo with projected texture.  We used to
demo this with several buffers worth of data as the head moved.  Didn't
seem to be a problem on the robots...kk

On 11/18/2010 2:29 PM, Josh Faust wrote:

> That change isn't actually going to help much with the Kinect, since
> there are relatively few invalid points.  I'm going to have to do some
> profiling to see what's up -- we haven't had a sensor that produces 300k
> points at 30hz before.
>
> Josh
>
> On Thu, Nov 18, 2010 at 2:12 PM, Radu Bogdan Rusu <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Dejan,
>
>     Just a bit of time until we make a release. :)
>
>     Cheers,
>     Radu.
>
>
>     On 11/18/2010 02:06 PM, Dejan Pangercic wrote:
>      > Hi Radu,
>      > what do we need to do to get the fix for rviz?
>      > D.
>      >
>      > On Thu, Nov 18, 2010 at 11:03 PM, Radu Bogdan Rusu
>      > <[hidden email] <mailto:[hidden email]>>  wrote:
>      >> This is a bug we just fixed in RViz.
>      >>
>      >> PointCloud2 data can contain NaN numbers, and that was hosing
>     things. RViz is now clever and displays the point clouds
>      >> without any issues.
>      >>
>      >> Alternatively, for a lightweight issue you can do:
>      >>
>      >> $ rosrun pcl_tutorials pointcloud_online_viewer
>     input_topic:=/your_topic
>      >>
>      >> to visualize PointCloud2 messages. No TF needed! Beautiful! :)
>      >>
>      >> Cheers,
>      >> Radu.
>      >>
>      >>
>      >> On 11/18/2010 01:53 PM, [hidden email]
>     <mailto:[hidden email]> wrote:
>      >>> My bag files are huge. It usually only takes ~20 seconds before it
>      >>> reaches the limit. Is there any way to decrease the amount of
>     points
>      >>> being recorded from the sensor? I know we can do some post
>     processing
>      >>> filtering with PCL but want to be able to make longer
>     recordings. Is
>      >>> anyone else having problems with RVIZ? I find that looking at the
>      >>> point cloud basically seizes up my machine.
>      >>>
>      >>> Thanks
>      >>> Ralph
>      >>>
>      >>> On Nov 18, 4:32 pm, Radu Bogdan Rusu<[hidden email]
>     <mailto:[hidden email]>>    wrote:
>      >>>> If you want to export them from BAGs into standalone PCD files
>     (one PointCloud per file) use:
>      >>>>
>      >>>> $ rosrun pcl_ros bag_to_pcd<bagfile> <topic> <output_directory>
>      >>>>
>      >>>> Cheers,
>      >>>> Radu.
>      >>>>
>      >>>> On 11/18/2010 01:09 PM, binarymillenium wrote:
>      >>>>
>      >>>>
>      >>>>
>      >>>>> There are these files in the ROS bag format:
>      >>>>
>      >>>>> https://code.ros.org/svn/data/trunk/ccny/
>      >>>>
>      >>>>> (and more infohttp://www.ros.org/wiki/kinect_node
>     <http://www.ros.org/wiki/kinect_node>)
>      >>>>
>      >>>>> -Lucas
>      >>>>
>      >>>>> On Thu, Nov 18, 2010 at 12:48 PM, yair
>     reshef<[hidden email]
>     <mailto:[hidden email]><mailto:[hidden email]
>     <mailto:[hidden email]>>>    wrote:
>      >>>>
>      >>>>>       missed your post *shame*
>      >>>>>       and so, i second
>      >>>>
>      >>>>>       On Thu, Nov 18, 2010 at 5:58 AM,
>     Ashwin<[hidden email]
>     <mailto:[hidden email]><mailto:[hidden email]
>     <mailto:[hidden email]>>>    wrote:
>      >>>>
>      >>>>>           Hi all,
>      >>>>
>      >>>>>              I was wondering if it would be a good idea to
>     make logs  of all the
>      >>>>>           raw data available from Kinect ( rgb and point
>     clouds.) If some has
>      >>>>>           integrated it with a mobile robot, any physical
>     state information
>      >>>>>           would be even better.
>      >>>>>              An immediate benefit of this would that even
>     people who haven't yet
>      >>>>>           got their hands on the device might still be able
>     to explore its
>      >>>>>           capabilities. It future it can also help to make
>     the testing phase and
>      >>>>>           also be a means to benchmarking different
>     approaches in common domain.
>      >>>>
>      >>>>>           ~Ashwin
>      >>>>
>      >>>>>       --
>      >>>>>       yair99@gmail
>      >>>>>       050-6301212
>      >>>>>       tlv, israel- Hide quoted text -
>      >>>>
>      >>>> - Show quoted text -
>      >> _______________________________________________
>      >> Ros-kinect mailing list
>      >> [hidden email] <mailto:[hidden email]>
>      >> https://code.ros.org/mailman/listinfo/ros-kinect
>      >>
>      >
>      >
>      >
>     _______________________________________________
>     Ros-kinect mailing list
>     [hidden email] <mailto:[hidden email]>
>     https://code.ros.org/mailman/listinfo/ros-kinect
>
>
>
>
> _______________________________________________
> Ros-kinect mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-kinect
_______________________________________________
Ros-kinect mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-kinect
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [Ros-kinect] Kinect Data Logging

Josh Faust
Narrow stereo with texture is ~15hz, not 30.

On Thu, Nov 18, 2010 at 11:30 PM, Kurt Konolige <[hidden email]> wrote:
Yes, you have - the narrow stereo with projected texture.  We used to
demo this with several buffers worth of data as the head moved.  Didn't
seem to be a problem on the robots...kk

On 11/18/2010 2:29 PM, Josh Faust wrote:
> That change isn't actually going to help much with the Kinect, since
> there are relatively few invalid points.  I'm going to have to do some
> profiling to see what's up -- we haven't had a sensor that produces 300k
> points at 30hz before.
>
> Josh
>
> On Thu, Nov 18, 2010 at 2:12 PM, Radu Bogdan Rusu <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Dejan,
>
>     Just a bit of time until we make a release. :)
>
>     Cheers,
>     Radu.
>
>
>     On 11/18/2010 02:06 PM, Dejan Pangercic wrote:
>      > Hi Radu,
>      > what do we need to do to get the fix for rviz?
>      > D.
>      >
>      > On Thu, Nov 18, 2010 at 11:03 PM, Radu Bogdan Rusu
>      > <[hidden email] <mailto:[hidden email]>>  wrote:
>      >> This is a bug we just fixed in RViz.
>      >>
>      >> PointCloud2 data can contain NaN numbers, and that was hosing
>     things. RViz is now clever and displays the point clouds
>      >> without any issues.
>      >>
>      >> Alternatively, for a lightweight issue you can do:
>      >>
>      >> $ rosrun pcl_tutorials pointcloud_online_viewer
>     input_topic:=/your_topic
>      >>
>      >> to visualize PointCloud2 messages. No TF needed! Beautiful! :)
>      >>
>      >> Cheers,
>      >> Radu.
>      >>
>      >>
>      >> On 11/18/2010 01:53 PM, [hidden email]
>     <mailto:[hidden email]> wrote:
>      >>> My bag files are huge. It usually only takes ~20 seconds before it
>      >>> reaches the limit. Is there any way to decrease the amount of
>     points
>      >>> being recorded from the sensor? I know we can do some post
>     processing
>      >>> filtering with PCL but want to be able to make longer
>     recordings. Is
>      >>> anyone else having problems with RVIZ? I find that looking at the
>      >>> point cloud basically seizes up my machine.
>      >>>
>      >>> Thanks
>      >>> Ralph
>      >>>
>      >>> On Nov 18, 4:32 pm, Radu Bogdan Rusu<[hidden email]
>     <mailto:[hidden email]>>    wrote:
>      >>>> If you want to export them from BAGs into standalone PCD files
>     (one PointCloud per file) use:
>      >>>>
>      >>>> $ rosrun pcl_ros bag_to_pcd<bagfile> <topic> <output_directory>
>      >>>>
>      >>>> Cheers,
>      >>>> Radu.
>      >>>>
>      >>>> On 11/18/2010 01:09 PM, binarymillenium wrote:
>      >>>>
>      >>>>
>      >>>>
>      >>>>> There are these files in the ROS bag format:
>      >>>>
>      >>>>> https://code.ros.org/svn/data/trunk/ccny/
>      >>>>
>      >>>>> (and more infohttp://www.ros.org/wiki/kinect_node
>     <http://www.ros.org/wiki/kinect_node>)
>      >>>>
>      >>>>> -Lucas
>      >>>>
>      >>>>> On Thu, Nov 18, 2010 at 12:48 PM, yair
>     reshef<[hidden email]
>     <mailto:[hidden email]><mailto:[hidden email]
>     <mailto:[hidden email]>>>    wrote:
>      >>>>
>      >>>>>       missed your post *shame*
>      >>>>>       and so, i second
>      >>>>
>      >>>>>       On Thu, Nov 18, 2010 at 5:58 AM,
>     Ashwin<[hidden email]
>     <mailto:[hidden email]><mailto:[hidden email]
>     <mailto:[hidden email]>>>    wrote:
>      >>>>
>      >>>>>           Hi all,
>      >>>>
>      >>>>>              I was wondering if it would be a good idea to
>     make logs  of all the
>      >>>>>           raw data available from Kinect ( rgb and point
>     clouds.) If some has
>      >>>>>           integrated it with a mobile robot, any physical
>     state information
>      >>>>>           would be even better.
>      >>>>>              An immediate benefit of this would that even
>     people who haven't yet
>      >>>>>           got their hands on the device might still be able
>     to explore its
>      >>>>>           capabilities. It future it can also help to make
>     the testing phase and
>      >>>>>           also be a means to benchmarking different
>     approaches in common domain.
>      >>>>
>      >>>>>           ~Ashwin
>      >>>>
>      >>>>>       --
>      >>>>>       yair99@gmail
>      >>>>>       050-6301212
>      >>>>>       tlv, israel- Hide quoted text -
>      >>>>
>      >>>> - Show quoted text -
>      >> _______________________________________________
>      >> Ros-kinect mailing list
>      >> [hidden email] <mailto:[hidden email]>
>      >> https://code.ros.org/mailman/listinfo/ros-kinect
>      >>
>      >
>      >
>      >
>     _______________________________________________
>     Ros-kinect mailing list
>     [hidden email] <mailto:[hidden email]>
>     https://code.ros.org/mailman/listinfo/ros-kinect
>
>
>
>
> _______________________________________________
> Ros-kinect mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-kinect
_______________________________________________
Ros-kinect mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-kinect


_______________________________________________
Ros-kinect mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-kinect
Loading...