I am working with students at UT Austin, developing vision code for an
autonomous vehicle. We want to use ROS, and need a 1394 Digital Camera
driver for use with the car. With a number of people involved, I am
looking for a stable, well-documented package with minimal
dependencies.

>From earlier messages on this ML, the recommended solution seems to be
the cameradc1394 package from the cmu-ros-pkg repository. I see that
the dcam driver recently went to the graveyard.

Is cameradc1394 still recommended?

That package does seem to work, but is not easy to install and
documentation is sketchy. It depends on libdc1394v2 in the
camera_drivers_experimental stack (no longer included in the ROS
installation), and on opencv2 (a bit heavy for a device driver). The
documentation is not linked into the ROS wiki and can be hard to
locate. The svn checkout command for cmu-rps-pkg listed on SourceForge
does not point to the repository trunk. The README in the top source
directory points to http://manipulation.programmingvision.com/, which
is not publicly accessible.

I understand that the WG systems do not use 1394 cameras. Without
hardware, it's hard to fully support such packages. The CMU developers
are no doubt busy with other activities, too. I certainly don't blame
anyone for the current situation.

Is there any prospect of publishing some 1394 camera driver on
code.ros.org with documentation linked into the ros.org/wiki?

I am not expert in digital cameras, but am willing to help. We have
several camera models for testing, and I can write documentation for
the ROS wiki. Just pulling all the information together in one place
would be good. But, I'd really like to see a fully supported package
with design and code reviews, test plans, etc. It's hard to provide
comprehensive robot vision support without this critical component.
-- 
 joq

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
ros-users mailing list
ros-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ros-users