The second pipeline is with vaapih264dec and vaapipostproc flags (which works on GPU but rendering is not fine), CPU is usage is very low as vaapipostproc works on GPU. And Browser shows proper processed feed.As we are passing the frame to Turbojpeg to encode to Browser, feed is continuous and core dump issues.Įxample Pipeline : rtspsrc latency=1 timeout=1000 ! decodebin ! videoconvert n-threads=4 ! capsfilter caps="video/x-raw,format=BGR" ! gvainference -gvatrack tracking-type=short-term name=gvainference ! queue silent=true max-size-buffers=1 leaky=upstream ! videoconvert ! queue silent=true max-size-buffers=1 leaky=upstream ! appsink max-buffers=1 drop=true We were working with two types of pipeline one is with decodebin and videoconvert flags (which works on CPU but rendering is absolutely fine), CPU is usage is very high as videoconvert works on CPU. Let me give you more details what we are trying to do.ġ)We are trying to get Render the processed feed through browser.Ģ)And Reduce the CPU usage by using VAAPI. Thank you and for prompt reply on question. gst/vaapi/gstvaapisink.c(461): gst_vaapisink_x11_handle_events (): /GstPipeline:pipeline0/GstAutoVideoSink:autovideosink0/GstVaapiSink:autovideosink0-actual-sink-vaapi Execution ended after 0:00:15.297272512 Setting pipeline to PAUSED. ERROR: from element /GstPipeline:pipeline0/GstAutoVideoSink:autovideosink0/GstVaapiSink:autovideosink0-actual-sink-vaapi: Output window was closed Additional debug info. Please run 'gdb gst-launch-1.0 20484' to continue debugging, Ctrl-C to quit, or Ctrl-\ to dump core. New clock: GstSystemClock Progress: (request) Sending PLAY request Progress: (request) Sending PLAY request Progress: (request) Sent PLAY request Redistribute latency. Got context from element 'autovideosink0-actual-sink-vaapi': =context, =(GstVaapiDisplay)"\(GstVaapiDisplayGLX\)\ vaapidisplayglx1" Progress: (open) Opening Stream Progress: (connect) Connecting to rtsp://10.0.0.232:8654/simulated_camera_0_1605255349463857408.sdp Progress: (open) Retrieving server options Progress: (open) Retrieving media info Progress: (request) SETUP stream 0 Progress: (open) Opened Stream Setting pipeline to PLAYING. Pipeline is live and does not need PREROLL. Recreate the error message then switch back to the Command Prompt (or Terminal) and review the messages on the screen.I can reproduce it by only added it to the gst-launch-1.0 rtspsrc location=rtsp://10.0.0.232:8654/simulated_camera_0_1605255349463857408.sdp ! queue ! rtpjitterbuffer latency=0 ! rtph264depay ! video/x-h264 ! vaapih264dec ! videoconvert ! video/x-raw ! gvapython module=/home/developer/Downloads/gvapython3_for_mliu.py ! autovideosink sync=false Setting pipeline to PAUSED. If you run macOS or Linux, locate the openshot-qt file and launch it from a Terminal session (in macOS the default path is /Applications/OpenShot Video Editor.app/Contents/MacOS/openshot-qt, in Linux the default path depends on how you downloaded the program). You can run this command by opening the Command Prompt then typing the command. If you run Windows, the command is usually C:\Program Files\OpenShot Video Editor\openshot-qt-cli.exe. You can also launch the application with the command-line interface to show more debug information on the screen. If you find the log files overwhelming, you can open a bug report and send us the openshot-qt.log and libopenshot.log files. Close OpenShot and open the log files for review. Then relaunch OpenShot, and recreate the errors. The best approach is to visit your C:\Users\(USERNAME)\.openshot_qt folder ( ~/.openshot_qt folder on macOS and Linux), delete the libopenshot.log and openshot-qt.log files. To help track down the issue, you can try using some of the debug tools. Unfortunately, the error message is on the same level as "An unknown error has occurred" in Windows. This is an error message from FFmpeg, the library that OpenShot uses to export videos.
0 Comments
Leave a Reply. |