FYI Kinect hacked driver for linux

classic Classic list List threaded Threaded
25 messages Options
12
Reply | Threaded
Open this post in threaded view
|

FYI Kinect hacked driver for linux

John Browning
Would be a great addition for a cheap ranging sensor

http://news.cnet.com/8301-13772_3-20022467-52.html
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
Very impressive sensor, especially considering the price.
Sample image I took today:
http://img809.imageshack.us/img809/5162/kinectscreenshot.png

Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
depth information.

2010/11/11 John Browning <[hidden email]>:

>
> Would be a great addition for a cheap ranging sensor
>
> http://news.cnet.com/8301-13772_3-20022467-52.html
> --
> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
> Sent from the ROS-Users mailing list archive at Nabble.com.
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

idryanov
Is anyone working on a ROS wrapper/driver for this? I have a Kinect
and I was thinking about starting to write one

On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
<[hidden email]> wrote:

> Very impressive sensor, especially considering the price.
> Sample image I took today:
> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>
> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
> depth information.
>
> 2010/11/11 John Browning <[hidden email]>:
>>
>> Would be a great addition for a cheap ranging sensor
>>
>> http://news.cnet.com/8301-13772_3-20022467-52.html
>> --
>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>> Sent from the ROS-Users mailing list archive at Nabble.com.
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

goncabrita
We're also considering doing a driver here at the lab, but we havent got a Kinect yet! Maybe next week :D

Gonçalo Cabrita
ISR - University of Coimbra
Portugal

On Thu, Nov 11, 2010 at 4:32 PM, Ivan Dryanovski <[hidden email]> wrote:
Is anyone working on a ROS wrapper/driver for this? I have a Kinect
and I was thinking about starting to write one

On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
<[hidden email]> wrote:
> Very impressive sensor, especially considering the price.
> Sample image I took today:
> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>
> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
> depth information.
>
> 2010/11/11 John Browning <[hidden email]>:
>>
>> Would be a great addition for a cheap ranging sensor
>>
>> http://news.cnet.com/8301-13772_3-20022467-52.html
>> --
>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>> Sent from the ROS-Users mailing list archive at Nabble.com.
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users


_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
In reply to this post by idryanov
I'm looking at publishing a PointCloud2. Have a lot of other stuff to
do though, so I might not be able to finish that node in the next
days. It's also very basic and probably a bit ugly, because I haven't
done much ROS coding outside of tutorials and testing of
orocos_toolchain_ros yet ;)

2010/11/11 Ivan Dryanovski <[hidden email]>:

> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
> and I was thinking about starting to write one
>
> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
> <[hidden email]> wrote:
>> Very impressive sensor, especially considering the price.
>> Sample image I took today:
>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>
>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>> depth information.
>>
>> 2010/11/11 John Browning <[hidden email]>:
>>>
>>> Would be a great addition for a cheap ranging sensor
>>>
>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>> --
>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Kurt Konolige
We'll probably do a prelim ROS driver as soon as we get a Kinect.

Cheers --Kurt

On Thu, Nov 11, 2010 at 9:34 AM, Stefan Kohlbrecher
<[hidden email]> wrote:

> I'm looking at publishing a PointCloud2. Have a lot of other stuff to
> do though, so I might not be able to finish that node in the next
> days. It's also very basic and probably a bit ugly, because I haven't
> done much ROS coding outside of tutorials and testing of
> orocos_toolchain_ros yet ;)
>
> 2010/11/11 Ivan Dryanovski <[hidden email]>:
>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>> and I was thinking about starting to write one
>>
>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>> <[hidden email]> wrote:
>>> Very impressive sensor, especially considering the price.
>>> Sample image I took today:
>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>
>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>> depth information.
>>>
>>> 2010/11/11 John Browning <[hidden email]>:
>>>>
>>>> Would be a great addition for a cheap ranging sensor
>>>>
>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>> --
>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Alex Bravo
In reply to this post by idryanov
Maybe we can start a bounty for releasing first ROS Kinect driver. ;)

On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:

> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
> and I was thinking about starting to write one
>
> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
> <[hidden email]> wrote:
>> Very impressive sensor, especially considering the price.
>> Sample image I took today:
>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>
>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>> depth information.
>>
>> 2010/11/11 John Browning <[hidden email]>:
>>>
>>> Would be a great addition for a cheap ranging sensor
>>>
>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>> --
>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Nicholas Butko
Does anyone working on this know how easy it will be to port the source code to other platforms (e.g. OSX)? How much of the code touches the OS in linux-specific ways?

Just out of curiosity.

--Nick

On Nov 11, 2010, at 10:18 AM, RobotNV wrote:

> Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
>
> On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
>
>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>> and I was thinking about starting to write one
>>
>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>> <[hidden email]> wrote:
>>> Very impressive sensor, especially considering the price.
>>> Sample image I took today:
>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>
>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>> depth information.
>>>
>>> 2010/11/11 John Browning <[hidden email]>:
>>>>
>>>> Would be a great addition for a cheap ranging sensor
>>>>
>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>> --
>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users

_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Lorenz Mösenlechner
It looks like the driver is based on libusb which means that it should
be pretty easy to port it to Windows and OSX. There are libusb
versions for almost all important operating systems.

Lorenz

> Does anyone working on this know how easy it will be to port the source code to other platforms (e.g. OSX)? How much of the code touches the OS in linux-specific ways?
>
> Just out of curiosity.
>
> --Nick
>
> On Nov 11, 2010, at 10:18 AM, RobotNV wrote:
>
> > Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
> >
> > On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
> >
> >> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
> >> and I was thinking about starting to write one
> >>
> >> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
> >> <[hidden email]> wrote:
> >>> Very impressive sensor, especially considering the price.
> >>> Sample image I took today:
> >>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
> >>>
> >>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
> >>> depth information.
> >>>
> >>> 2010/11/11 John Browning <[hidden email]>:
> >>>>
> >>>> Would be a great addition for a cheap ranging sensor
> >>>>
> >>>> http://news.cnet.com/8301-13772_3-20022467-52.html
> >>> _______________________________________________
> >>> ros-users mailing list
> >>> [hidden email]
> >>> https://code.ros.org/mailman/listinfo/ros-users
> >>>
> >> _______________________________________________
> >> ros-users mailing list
> >> [hidden email]
> >> https://code.ros.org/mailman/listinfo/ros-users
> > _______________________________________________
> > ros-users mailing list
> > [hidden email]
> > https://code.ros.org/mailman/listinfo/ros-users
>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>

--
Lorenz Mösenlechner            | [hidden email]
Technische Universität München | Boltzmannstr. 3
85748 Garching bei München     | Germany
http://ias.cs.tum.edu/         | Tel: +49 (89) 289-26910
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
In reply to this post by Alex Bravo
Mine currently looks like that:
http://img837.imageshack.us/img837/3792/kinectrviz.png
Obviously a little bug with the point cloud indices. The interesting
thing is probably to get calibration and synchronization of depth and
RGB data right. So far the driver gets seperate images for both.

2010/11/11 RobotNV <[hidden email]>:

> Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
>
> On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
>
>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>> and I was thinking about starting to write one
>>
>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>> <[hidden email]> wrote:
>>> Very impressive sensor, especially considering the price.
>>> Sample image I took today:
>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>
>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>> depth information.
>>>
>>> 2010/11/11 John Browning <[hidden email]>:
>>>>
>>>> Would be a great addition for a cheap ranging sensor
>>>>
>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>> --
>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
In reply to this post by Nicholas Butko
There are people running it on OSX, there is a thread in the
http://groups.google.com/group/openkinect group on that. One of guys
there posted this video:
http://vimeo.com/16734124

2010/11/11 Nicholas Butko <[hidden email]>:

> Does anyone working on this know how easy it will be to port the source code to other platforms (e.g. OSX)? How much of the code touches the OS in linux-specific ways?
>
> Just out of curiosity.
>
> --Nick
>
> On Nov 11, 2010, at 10:18 AM, RobotNV wrote:
>
>> Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
>>
>> On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
>>
>>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>>> and I was thinking about starting to write one
>>>
>>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>>> <[hidden email]> wrote:
>>>> Very impressive sensor, especially considering the price.
>>>> Sample image I took today:
>>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>>
>>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>>> depth information.
>>>>
>>>> 2010/11/11 John Browning <[hidden email]>:
>>>>>
>>>>> Would be a great addition for a cheap ranging sensor
>>>>>
>>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>>> --
>>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>>> _______________________________________________
>>>>> ros-users mailing list
>>>>> [hidden email]
>>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>>
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

idryanov
In reply to this post by Stefan Kohlbrecher
Stefan,

Are you working with Hector Martin's driver?



On Thu, Nov 11, 2010 at 1:46 PM, Stefan Kohlbrecher
<[hidden email]> wrote:

> Mine currently looks like that:
> http://img837.imageshack.us/img837/3792/kinectrviz.png
> Obviously a little bug with the point cloud indices. The interesting
> thing is probably to get calibration and synchronization of depth and
> RGB data right. So far the driver gets seperate images for both.
>
> 2010/11/11 RobotNV <[hidden email]>:
>> Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
>>
>> On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
>>
>>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>>> and I was thinking about starting to write one
>>>
>>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>>> <[hidden email]> wrote:
>>>> Very impressive sensor, especially considering the price.
>>>> Sample image I took today:
>>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>>
>>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>>> depth information.
>>>>
>>>> 2010/11/11 John Browning <[hidden email]>:
>>>>>
>>>>> Would be a great addition for a cheap ranging sensor
>>>>>
>>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>>> --
>>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>>> _______________________________________________
>>>>> ros-users mailing list
>>>>> [hidden email]
>>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>>
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
Yes, I basically took the glview code and added a PointCloud2
publisher that publishes at the end of the depthimg(uint16_t *buf, int
width, int height) callback. I think Hector Martin is also working on
further optimizing the USB code and the whole driver. I think it's the
only publicly available driver, isn't it?

2010/11/11 Ivan Dryanovski <[hidden email]>:

> Stefan,
>
> Are you working with Hector Martin's driver?
>
>
>
> On Thu, Nov 11, 2010 at 1:46 PM, Stefan Kohlbrecher
> <[hidden email]> wrote:
>> Mine currently looks like that:
>> http://img837.imageshack.us/img837/3792/kinectrviz.png
>> Obviously a little bug with the point cloud indices. The interesting
>> thing is probably to get calibration and synchronization of depth and
>> RGB data right. So far the driver gets seperate images for both.
>>
>> 2010/11/11 RobotNV <[hidden email]>:
>>> Maybe we can start a bounty for releasing first ROS Kinect driver. ;)
>>>
>>> On Nov 11, 2010, at 8:32 AM, Ivan Dryanovski <[hidden email]> wrote:
>>>
>>>> Is anyone working on a ROS wrapper/driver for this? I have a Kinect
>>>> and I was thinking about starting to write one
>>>>
>>>> On Thu, Nov 11, 2010 at 10:22 AM, Stefan Kohlbrecher
>>>> <[hidden email]> wrote:
>>>>> Very impressive sensor, especially considering the price.
>>>>> Sample image I took today:
>>>>> http://img809.imageshack.us/img809/5162/kinectscreenshot.png
>>>>>
>>>>> Minimum range is about 0.5m, from 1 to 4 meters+ one gets very dense
>>>>> depth information.
>>>>>
>>>>> 2010/11/11 John Browning <[hidden email]>:
>>>>>>
>>>>>> Would be a great addition for a cheap ranging sensor
>>>>>>
>>>>>> http://news.cnet.com/8301-13772_3-20022467-52.html
>>>>>> --
>>>>>> View this message in context: http://ros-users.122217.n3.nabble.com/FYI-Kinect-hacked-driver-for-linux-tp1882853p1882853.html
>>>>>> Sent from the ROS-Users mailing list archive at Nabble.com.
>>>>>> _______________________________________________
>>>>>> ros-users mailing list
>>>>>> [hidden email]
>>>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>>>
>>>>> _______________________________________________
>>>>> ros-users mailing list
>>>>> [hidden email]
>>>>> https://code.ros.org/mailman/listinfo/ros-users
>>>>>
>>>> _______________________________________________
>>>> ros-users mailing list
>>>> [hidden email]
>>>> https://code.ros.org/mailman/listinfo/ros-users
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>>
>> _______________________________________________
>> ros-users mailing list
>> [hidden email]
>> https://code.ros.org/mailman/listinfo/ros-users
>>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

idryanov
> Yes, I basically took the glview code and added a PointCloud2
> publisher that publishes at the end of the depthimg(uint16_t *buf, int
> width, int height) callback. I think Hector Martin is also working on
> further optimizing the USB code and the whole driver. I think it's the
> only publicly available driver, isn't it?


I believe so. I've been poking around the code and the forums, and I
have some questions. Maybe you have some insight:

1. Is the depth image really 640x480, or is that oversampled? The
wikipedia page states the depth sensor has an resolution of 320×240
pixels. If it's oversampled, where does that take place - in the
driver, or the device itself? I prefer not inflating the point cloud
with oversampled data

2. What is the relationship between the values in the depth_frame and
the distance in meters? It doesn't appear to be linear

3. I read somewhere the device's range can be set dynamically. I'm
guessing one of the inits in inits.c could be responsible for the
range.
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Hordur Johannsson
The Kinetic is based on technology from Primesense. They have a spec on their website http://www.primesense.com/files/FMF_2.PDF
It works by creating stereo with structured light and a single camera. You get a 640x480 resolution range image. The range resolution is 1cm and the range is 0.8-3.5m.
It can operate upto 60fps.

This is all just based on the spec, I haven't used this sensor before.

Hordur

On Thu, Nov 11, 2010 at 3:21 PM, Ivan Dryanovski <[hidden email]> wrote:
> Yes, I basically took the glview code and added a PointCloud2
> publisher that publishes at the end of the depthimg(uint16_t *buf, int
> width, int height) callback. I think Hector Martin is also working on
> further optimizing the USB code and the whole driver. I think it's the
> only publicly available driver, isn't it?


I believe so. I've been poking around the code and the forums, and I
have some questions. Maybe you have some insight:

1. Is the depth image really 640x480, or is that oversampled? The
wikipedia page states the depth sensor has an resolution of 320×240
pixels. If it's oversampled, where does that take place - in the
driver, or the device itself? I prefer not inflating the point cloud
with oversampled data

2. What is the relationship between the values in the depth_frame and
the distance in meters? It doesn't appear to be linear

3. I read somewhere the device's range can be set dynamically. I'm
guessing one of the inits in inits.c could be responsible for the
range.
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users


_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stéphane Magnenat
In reply to this post by Stefan Kohlbrecher
Hello,

I wrote a ROS driver [1] for the kinect based on libfreenect [2]. The driver somehow works, but libfreenect is currently far from being finished, for instance I had to fix a memory trash there. Of course, all contributions are welcome to improve this driver :-)

All the best,

Stéphane

[1] https://github.com/ethz-asl/ros-drivers/tree/master/microsoft_kinect/
[2] http://git.marcansoft.com/?p=libfreenect.git
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
In reply to this post by idryanov
> 1. Is the depth image really 640x480, or is that oversampled? The
> wikipedia page states the depth sensor has an resolution of 320×240
> pixels. If it's oversampled, where does that take place - in the
> driver, or the device itself? I prefer not inflating the point cloud
> with oversampled data
I think the device itself reports the data with this size. If you look
at the picture I posted in the second post you also see that there are
for example one pixel sized holes in the 640x480 sized depth image,
which should not exist if some very simple interpolation scheme would
be used to blow up a 320x240 image to 640x480.
From what I read beforehand, the original Project Natal was supposed
to be 640x480, then Microsoft reportedly "downgraded" to 320x240 for
cost reasons (see
http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=kinect+downgraded+to+320x240).
Now the sensor appears to deliver 640x480 again, which might or might
not be just blown up 320x240on the onboard ASIC.

> 2. What is the relationship between the values in the depth_frame and
> the distance in meters? It doesn't appear to be linear
That´s really the interesting question, along with others like how to
calibrate visual and depth image to get real RGB-D data. With the
current state of affairs one can generate some impressive looking
images, but to leverage the full potential of the sensor these
calibration questions really have to be solved.

> 3. I read somewhere the device's range can be set dynamically. I'm
> guessing one of the inits in inits.c could be responsible for the
> range.
That´s more stuff that will probably be discovered in the coming
days/weeks. Still very impressive how good the sensor works already
right out of the box.

> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Alex Trevor
Wow, looks like quite a few people are working on kinect nodes; we should probably combine efforts!  We’ve also been working on a kinect node for ROS based on Hector Martin’s driver, available at:
git://github.com/atrevor/kinect_node.git

Short youtube rviz screencap:
http://www.youtube.com/watch?v=IxRIL1izvDs

We assume that you have libfreenect from Hector Martin’s repo in the same parent directory (../libfreenect w.r.t. kinect_node):
git://git.marcansoft.com/libfreenect.git

So far, it publishes the camera image, as well as a PointCloud2 of PointXYZRGBs.  The RGB camera’s image is projected onto the range data, resulting in a color point cloud.  We calibrated the RGB camera, but haven’t yet calibrated the range camera -- since we can’t just use our normal checkerboard calibration target for this :) .  As Ivan and Stefan noted, the ranges we get are a little odd -- they definitely don't seem to be linear.  We calibrated it so the range is approximately correct at 2m for our sensor, but I agree that we really need to know how these work to make much progress. We'll probably do some testing shortly with targets at various ranges to attempt to address this.  Any input would be greatly appreciated!

On Thu, Nov 11, 2010 at 5:41 PM, Stefan Kohlbrecher <[hidden email]> wrote:
> 1. Is the depth image really 640x480, or is that oversampled? The
> wikipedia page states the depth sensor has an resolution of 320×240
> pixels. If it's oversampled, where does that take place - in the
> driver, or the device itself? I prefer not inflating the point cloud
> with oversampled data
I think the device itself reports the data with this size. If you look
at the picture I posted in the second post you also see that there are
for example one pixel sized holes in the 640x480 sized depth image,
which should not exist if some very simple interpolation scheme would
be used to blow up a 320x240 image to 640x480.
From what I read beforehand, the original Project Natal was supposed
to be 640x480, then Microsoft reportedly "downgraded" to 320x240 for
cost reasons (see
http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=kinect+downgraded+to+320x240).
Now the sensor appears to deliver 640x480 again, which might or might
not be just blown up 320x240on the onboard ASIC.

> 2. What is the relationship between the values in the depth_frame and
> the distance in meters? It doesn't appear to be linear
That´s really the interesting question, along with others like how to
calibrate visual and depth image to get real RGB-D data. With the
current state of affairs one can generate some impressive looking
images, but to leverage the full potential of the sensor these
calibration questions really have to be solved.

> 3. I read somewhere the device's range can be set dynamically. I'm
> guessing one of the inits in inits.c could be responsible for the
> range.
That´s more stuff that will probably be discovered in the coming
days/weeks. Still very impressive how good the sensor works already
right out of the box.

> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users


_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Alex Trevor
Apologies for the broken github link, the correct one is:

On Fri, Nov 12, 2010 at 2:20 AM, Alex Trevor <[hidden email]> wrote:
Wow, looks like quite a few people are working on kinect nodes; we should probably combine efforts!  We’ve also been working on a kinect node for ROS based on Hector Martin’s driver, available at:
git://github.com/atrevor/kinect_node.git

Short youtube rviz screencap:
http://www.youtube.com/watch?v=IxRIL1izvDs

We assume that you have libfreenect from Hector Martin’s repo in the same parent directory (../libfreenect w.r.t. kinect_node):
git://git.marcansoft.com/libfreenect.git

So far, it publishes the camera image, as well as a PointCloud2 of PointXYZRGBs.  The RGB camera’s image is projected onto the range data, resulting in a color point cloud.  We calibrated the RGB camera, but haven’t yet calibrated the range camera -- since we can’t just use our normal checkerboard calibration target for this :) .  As Ivan and Stefan noted, the ranges we get are a little odd -- they definitely don't seem to be linear.  We calibrated it so the range is approximately correct at 2m for our sensor, but I agree that we really need to know how these work to make much progress. We'll probably do some testing shortly with targets at various ranges to attempt to address this.  Any input would be greatly appreciated!

On Thu, Nov 11, 2010 at 5:41 PM, Stefan Kohlbrecher <[hidden email]> wrote:
> 1. Is the depth image really 640x480, or is that oversampled? The
> wikipedia page states the depth sensor has an resolution of 320×240
> pixels. If it's oversampled, where does that take place - in the
> driver, or the device itself? I prefer not inflating the point cloud
> with oversampled data
I think the device itself reports the data with this size. If you look
at the picture I posted in the second post you also see that there are
for example one pixel sized holes in the 640x480 sized depth image,
which should not exist if some very simple interpolation scheme would
be used to blow up a 320x240 image to 640x480.
From what I read beforehand, the original Project Natal was supposed
to be 640x480, then Microsoft reportedly "downgraded" to 320x240 for
cost reasons (see
http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=kinect+downgraded+to+320x240).
Now the sensor appears to deliver 640x480 again, which might or might
not be just blown up 320x240on the onboard ASIC.

> 2. What is the relationship between the values in the depth_frame and
> the distance in meters? It doesn't appear to be linear
That´s really the interesting question, along with others like how to
calibrate visual and depth image to get real RGB-D data. With the
current state of affairs one can generate some impressive looking
images, but to leverage the full potential of the sensor these
calibration questions really have to be solved.

> 3. I read somewhere the device's range can be set dynamically. I'm
> guessing one of the inits in inits.c could be responsible for the
> range.
That´s more stuff that will probably be discovered in the coming
days/weeks. Still very impressive how good the sensor works already
right out of the box.

> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users



_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|

Re: FYI Kinect hacked driver for linux

Stefan Kohlbrecher
Nice, that already looks pretty good. Do you have the same ~4cm
vertical offset between color and depth data?
http://img717.imageshack.us/img717/3955/kinectrviztrevor.png
It's pretty visible when looking at the screen in the lower portion.
I'm well aware that this is basically uncalibrated, my question is
more directed towards the variability of alignment between different
Kinect devices.

2010/11/12 Alex Trevor <[hidden email]>:

> Apologies for the broken github link, the correct one is:
> https://github.com/atrevor/kinect_node.git
> On Fri, Nov 12, 2010 at 2:20 AM, Alex Trevor <[hidden email]> wrote:
>>
>> Wow, looks like quite a few people are working on kinect nodes; we should
>> probably combine efforts!  We’ve also been working on a kinect node for ROS
>> based on Hector Martin’s driver, available at:
>> git://github.com/atrevor/kinect_node.git
>>
>> Short youtube rviz screencap:
>> http://www.youtube.com/watch?v=IxRIL1izvDs
>>
>> We assume that you have libfreenect from Hector Martin’s repo in the same
>> parent directory (../libfreenect w.r.t. kinect_node):
>> git://git.marcansoft.com/libfreenect.git
>>
>> So far, it publishes the camera image, as well as a PointCloud2 of
>> PointXYZRGBs.  The RGB camera’s image is projected onto the range data,
>> resulting in a color point cloud.  We calibrated the RGB camera, but haven’t
>> yet calibrated the range camera -- since we can’t just use our normal
>> checkerboard calibration target for this :) .  As Ivan and Stefan noted, the
>> ranges we get are a little odd -- they definitely don't seem to be linear.
>>  We calibrated it so the range is approximately correct at 2m for our
>> sensor, but I agree that we really need to know how these work to make much
>> progress. We'll probably do some testing shortly with targets at various
>> ranges to attempt to address this.  Any input would be greatly appreciated!
>> On Thu, Nov 11, 2010 at 5:41 PM, Stefan Kohlbrecher
>> <[hidden email]> wrote:
>>>
>>> > 1. Is the depth image really 640x480, or is that oversampled? The
>>> > wikipedia page states the depth sensor has an resolution of 320×240
>>> > pixels. If it's oversampled, where does that take place - in the
>>> > driver, or the device itself? I prefer not inflating the point cloud
>>> > with oversampled data
>>> I think the device itself reports the data with this size. If you look
>>> at the picture I posted in the second post you also see that there are
>>> for example one pixel sized holes in the 640x480 sized depth image,
>>> which should not exist if some very simple interpolation scheme would
>>> be used to blow up a 320x240 image to 640x480.
>>> From what I read beforehand, the original Project Natal was supposed
>>> to be 640x480, then Microsoft reportedly "downgraded" to 320x240 for
>>> cost reasons (see
>>>
>>> http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=kinect+downgraded+to+320x240).
>>> Now the sensor appears to deliver 640x480 again, which might or might
>>> not be just blown up 320x240on the onboard ASIC.
>>>
>>> > 2. What is the relationship between the values in the depth_frame and
>>> > the distance in meters? It doesn't appear to be linear
>>> That´s really the interesting question, along with others like how to
>>> calibrate visual and depth image to get real RGB-D data. With the
>>> current state of affairs one can generate some impressive looking
>>> images, but to leverage the full potential of the sensor these
>>> calibration questions really have to be solved.
>>>
>>> > 3. I read somewhere the device's range can be set dynamically. I'm
>>> > guessing one of the inits in inits.c could be responsible for the
>>> > range.
>>> That´s more stuff that will probably be discovered in the coming
>>> days/weeks. Still very impressive how good the sensor works already
>>> right out of the box.
>>>
>>> > _______________________________________________
>>> > ros-users mailing list
>>> > [hidden email]
>>> > https://code.ros.org/mailman/listinfo/ros-users
>>> >
>>> _______________________________________________
>>> ros-users mailing list
>>> [hidden email]
>>> https://code.ros.org/mailman/listinfo/ros-users
>>
>
>
> _______________________________________________
> ros-users mailing list
> [hidden email]
> https://code.ros.org/mailman/listinfo/ros-users
>
>
_______________________________________________
ros-users mailing list
[hidden email]
https://code.ros.org/mailman/listinfo/ros-users
12