Closed dorp92 closed 1 year ago
tschoonj
commented
1 year ago nvjpegenc seems to be part of an external package, provided by nvidia, and is not provided by any of the GNOME gtreamer packages.
So if you want to use this through your conda installation, you should create a conda package for it first.
dorp92
commented
1 year ago Hi, Thanks for the answer. What I am looking for is how to make the gstreamer package installed by conda to recognize external plugins. Before making conda package for it I want to make sure the plugin works at all.
dorp92
commented
1 year ago I don't understand why you closed that issue, I believe exernal plugin support is an importent usecase.
tschoonj
commented
1 year ago If you want to use this particular plugin in a neat way within a conda environment, you will need to write a dedicated conda recipe for it, and you may want to submit it to conda-forge as it may be useful for others.
Solution to issue cannot be found in the documentation.
Issue
I am trying to run nvidia jetbot 4.6.1 camera example https://github.com/NVIDIA-AI-IOT/jetbot/blob/master/notebooks/teleoperation/teleoperation.ipynb on jetson nano. I was able to run it inside the provided docker by nvidia. But when I try to build my own conda enviorment outside docker nvjpegenc is missing. I tried to install servral gst plugins but nothing seems to work Maybe its path issue? How do I install spesific plugin (like https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvjpegenc.html) inside the same conda enviorment as gstreamer? Spent few days on this issue, Thanks alot for the help.
`Error Traceback (most recent call last) Cell In[1], line 2 1 import ipywidgets.widgets as widgets ----> 2 from jetbot import bgr8_to_jpeg 3 from jetbot import Camera 4 image = widgets.Image(format='jpeg', width=300, height=300)
File ~/micromamba/envs/dev/lib/python3.9/site-packages/jetbot-0.4.3-py3.9.egg/jetbot/init.py:5 3 from .motor import Motor 4 from .robot import Robot ----> 5 from .image import bgr8_to_jpeg 6 from .object_detection import ObjectDetector
File ~/micromamba/envs/dev/lib/python3.9/site-packages/jetbot-0.4.3-py3.9.egg/jetbot/image.py:6 2 import cv2 3 from .jpeg_encoder import JpegEncoder ----> 6 _ENCODER = JpegEncoder(width=224, height=224, fps=21) 9 def bgr8_to_jpeg_gst(value): 10 return _ENCODER.encode(value)
File ~/micromamba/envs/dev/lib/python3.9/site-packages/jetbot-0.4.3-py3.9.egg/jetbot/jpeg_encoder.py:34, in JpegEncoder.init(self, width, height, fps) 19 CAPS = "video/x-raw,format=BGR,width={width},height={height},framerate={fps}/1".format( 20 width=width, 21 height=height, 22 fps=fps 23 ) 25 GST_STRING = 'appsrc name=src emit-signals=True is-live=True caps=video/x-raw,format=BGR,width={width},height={height},framerate={fps}/1 !'\ 26 ' nvjpegenc '\ 27 '! image/jpeg,width={width},height={height},framerate={fps}/1 !'\ (...) 31 fps=fps 32 ) ---> 34 self.pipeline = Gst.parse_launch(GST_STRING) 35 self.appsrc = self.pipeline.get_by_name('src') 37 appsink = self.pipeline.get_by_name('sink')
Error: gst_parse_error: no element "nvjpegenc" (1)`
Installed packages
Environment info