NVIDIA Jetson Xavier - Video Capture and Display using GStreamer

From RidgeRun Developer Wiki



Previous: Video Capture and Display/Software Support Index Next: Video Capture and Display/Software Support/Libargus








GStreamer is a framework for creating streaming media applications. The fundamental design comes from the video pipeline at the Oregon Graduate Institute, as well as some ideas from DirectShow.

GStreamer Capture

GStreamer provides different commands for capturing images where two are nvarguscamerasrc and v4l2src.

nvarguscamerasrc

nvarguscamera src is used when the camera generates images of the Bayer format because it uses the ISP to change the images to a visible format.

ISP.png
Figure 1. ISP Bayer to YUV



nvarguscamerasrc road
Figure 2. nvarguscamerasrc road

v4l2src

v4l2src captures the images in the format provided by the camera, and is used when the camera uses visible formats.

v4l2src road
Figure 3. v4l2src road

GStreamer Display

GStreamer provides different commands for displaying images where two are nvoverlaysink and xvimagesink.

nvoverlaysink

nvoverlaysink is used when you need to use NVMM memory but also could use normal memory, this command writes directly into the screen buffer.

xvimagesink

xvimagesink is used when you need to use normal memory, this command opens a screen with the output

Pipeline Examples

For a complete summary of the samples go to https://developer.ridgerun.com/wiki/index.php?title=Xavier/GStreamer_Pipelines



Previous: Video Capture and Display/Software Support Index Next: Video Capture and Display/Software Support/Libargus