Image Stitching for NVIDIA Jetson/Examples/Using Gstd: Difference between revisions
(Add header) |
mNo edit summary |
||
| Line 1: | Line 1: | ||
<noinclude> | <noinclude> | ||
{{Image_Stitching_for_NVIDIA_Jetson/Head | Image Stitching, CUDA, Stitcher, Panorama}} | {{Image_Stitching_for_NVIDIA_Jetson/Head | }} | ||
</noinclude> | |||
<noinclude> | |||
{{Image_Stitching_for_NVIDIA_Jetson/Head|previous=Examples/Basic pipelines|next=Examples/Nano,TX1,TX2,Xavier_Pipelines|keywords=Image Stitching, Image Stitching, CUDA, Stitcher, Panorama}} | |||
</noinclude> | </noinclude> | ||
[[GStreamer_Daemon]] is a gst-launch on steroids where you can create a GStreamer pipeline, play, pause, change speed, skip around, and even change element parameter settings all while the pipeline is active. | [[GStreamer_Daemon]] is a gst-launch on steroids where you can create a GStreamer pipeline, play, pause, change speed, skip around, and even change element parameter settings all while the pipeline is active. | ||
| Line 23: | Line 28: | ||
gstd-client pipeline_play stitcher_pipeline | gstd-client pipeline_play stitcher_pipeline | ||
</syntaxhighlight> | </syntaxhighlight> | ||
<noinclude> | |||
{{Image_Stitching_for_NVIDIA_Jetson/Foot|Examples/Basic pipelines|Examples/Nano,TX1,TX2,Xavier_Pipelines}} | |||
</noinclude> | |||
Revision as of 18:19, 17 August 2020
| Image Stitching for NVIDIA®Jetson™ |
|---|
 |
| Before Starting |
| Image Stitching Basics |
| Overview |
| Getting Started |
| User Guide |
| Resources |
| Examples |
| Spherical Video |
| Performance |
| Contact Us |
|
| Image Stitching for NVIDIA®Jetson™ |
|---|
|
| Before Starting |
| Image Stitching Basics |
| Overview |
| Getting Started |
| User Guide |
| Resources |
| Examples |
| Spherical Video |
| Performance |
| Contact Us |
|
|
GStreamer_Daemon is a gst-launch on steroids where you can create a GStreamer pipeline, play, pause, change speed, skip around, and even change element parameter settings all while the pipeline is active.
Running with bash GSTD client
The main difference in the pipeline syntax is that the homographies now need to be surrounded with escaped single quotes:
gstd # Launch gstd daemon, this will separate itself into its own process
gstd-client pipeline_create stitcher_pipeline \
cudastitcher name=cuda \
left-center-homography=\'{\"h00\":2.47795806e-01, \"h01\":1.83125651e-01, \"h02\":1511, \"h10\":-9.58951851e-02, \"h11\":1.00466096e+00, \"h12\":3.43046193e+01, \"h20\":-1.84997102e-04, \"h21\":4.48703017e-05, \"h22\":1}\' \
right-center-homography=\'{\"h00\":1.07033034e+00, \"h01\":-8.91500609e-02, \"h02\":-1504, \"h10\":2.06014232e-02, \"h11\":1.06485384e+00, \"h12\":-2.68451772e+01, \"h20\":3.98503995e-06, \"h21\":4.20860985e-05, \"h22\":1}\' \
nvarguscamerasrc maxperf=true sensor-id=0 ! nvvidconv ! "video/x-raw(memory:NVMM), width=1920, height=1080" ! queue ! cuda.sink_0 \
nvarguscamerasrc maxperf=true sensor-id=1 ! nvvidconv ! "video/x-raw(memory:NVMM), width=1920, height=1080" ! queue ! cuda.sink_1 \
nvarguscamerasrc maxperf=true sensor-id=2 ! nvvidconv ! "video/x-raw(memory:NVMM), width=1920, height=1080" ! queue ! cuda.sink_2 \
cuda. ! perf print-arm-load=true ! queue ! nvvidconv ! "video/x-raw(memory:NVMM), width=1920, height=360" ! nvoverlaysink
gstd-client pipeline_play stitcher_pipeline