KINOVA
®
Gen3 Ultra lightweight robot
User Guide
130
Requirements:
• GStreamer version: 1.8.3 and above
• Supported operating systems:
º Windows 7 and later
º Ubuntu 16.04 and later
Using GStreamer
GStreamer pipelines can be called from the command line using the
gst-launch-1.0 utility
.
For integration with applications, GStreamer offers a number of
language bindings
, which
include both C++ and Python.
Official documentation for the GStreamer framework, including installation instructions,
application development guidance, and tutorials can be accessed here:
https://
gstreamer.freedesktop.org/documentation/
Windows command examples
This section provides examples of using GStreamer with the robot vision module camera
streams on the Windows command line.
Color stream CLI example: command to display the color stream
gst-launch-1.0.exe rtspsrc location=rtsp://192.168.1.10/color
latency=30 ! rtph264depay ! avdec_h264 ! autovideosink
Unpacking the example:
•
gst-launch-1.0.exe
: launch GStreamer
•
rtspsrc location=rtsp://192.168.1.10/color latency=30
: connect to the
RTSP server at the color stream URL with latency of 30 ms.
•
rtph264depay
: Extract H.264 video payload from RTP packets
•
avdec_h264
: decode H.264 video
•
autovideosink
: search computer registry for video sink (player) and plays decoded video
stream.
Depth stream CLI example: command to display the depth stream
gst-launch-1.0.exe rtspsrc location=rtsp://192.168.1.10/depth
latency=30 ! rtpgstdepay ! videoconvert ! autovideosink
Unpacking the example:
•
gst-launch-1.0.exe
: launch GStreamer
•
rtspsrc location=rtsp://192.168.1.10/depth latency=30
: connect to the
RTSP server at the color stream URL with latency of 30 ms.
•
rtpgstdepay
: extract GStreamer buffers from RTP packets
•
videoconvert
: automatically convert the video to a format understandable to the chosen
video sink in the next step
•
autovideosink
: search computer registry for video sink (player) and plays decoded video
stream.
Working with camera streams using GStreamer
This section describes how to work with vision module camera streams using the
GStreamer framework.
GStreamer
Kinova recommends that developers use the
GStreamer
framework for handling the camera's
sensor streams. GStreamer offers a pipeline-based framework that allows you to link together
plugins for image-processing workflows.