There are a lot of different USB camera drivers out there building on separate underlying libraries (gscam, uvc_cam, usb_cam). I Heart Robotics had a nice review ~6 months ago [1]. A USB camera driver would have to follow the same path that Jack O'Quin pioneered. We would need someone to step forward to survey the current field of USB camera drivers, choose the one(s) worth stabilizing, develop where appropriate, review, and release. As before, we're happy to support the process, but, as a policy, we don't develop drivers for hardware we don't use. The camera1394 review [2] should give a general idea of what to expect with the process. Now that the ROS camera API is relatively stable, we can start dissolving the idea of the monolithic 'camera_drivers' stack. This means that a new USB camera driver would not have to meet the camera_drivers freeze date -- it would likely be in its own separate stack anyway. - Ken [1]: http://www.iheartrobotics.com/2010/05/testing-ros-usb-camera-drivers.html [2]: http://www.ros.org/wiki/camera1394/Reviews/2010-05-27_API_Review On Sat, Dec 4, 2010 at 7:55 PM, Bill Morris wrote: > Is there any chance we can get a nodelet based USB camera driver in for > Diamondback? > > On Sat, 2010-12-04 at 19:45 -0800, Ken Conley wrote: >> It's currently sidetracked because of our focus on the Kinect.  We >> figured it would be a shame if people had to wait for E Turtle for >> good Kinect support. We hope to still get it in for Diamondback, and >> we've relaxed the freeze date on Kinect-affected stacks. >> >>  - Ken >> >> On Sat, Dec 4, 2010 at 7:26 PM, Bill Morris wrote: >> > Does anyone know the status of the nodelet based image pipeline? >> > >> > The roadmap seems incomplete. >> > http://www.ros.org/wiki/image_pipeline/Roadmap >> > >> > I'm most interested in the status of a nodelet based image_view. >> > >> > _______________________________________________ >> > ros-users mailing list >> > ros-users@code.ros.org >> > https://code.ros.org/mailman/listinfo/ros-users >> > > > >