Please provide complete information as applicable to your setup.
cuda arch 7.2.2
deepstream-app version 6.0.1
I need to add a queue after streammux, get the real-time latitude and longitude in this queue and add the latitude and longitude to the data of the video frame, to ensure that when the recognition result is obtained at the end, the obtained latitude and longitude is the relatively accurate latitude and longitude of the current frame
Is the “latitude and longitude” for the whole frame or for some objects? You can add the "“latitude and longitude” by customizing your own NVIDIA DeepStream SDK API Reference: _NvDsUserMeta Struct Reference | NVIDIA Docs with NVIDIA DeepStream SDK API Reference: _NvDsFrameMeta Struct Reference | NVIDIA Docs or NVIDIA DeepStream SDK API Reference: _NvDsObjectMeta Struct Reference | NVIDIA Docs according to your scenario.
There is NvDsUserMeta customization sample in /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-gst-metadata-test
latitude and longitude” for the whole frame
Does it mean that you can only write c++ classes first and then call them on python, and you can’t add them directly through python
Python is just a script language, the binary under it can be generated by c/c++ or any other language. You can customize user meta binding. deepstream_python_apps/CUSTOMUSERMETAGUIDE.md at master · NVIDIA-AI-IOT/deepstream_python_apps · GitHub
When I use the camera, the frame_num and batch_id obtained through the frame meta are always 0. Is this because of this design, or is it because my configuration is wrong?
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
Where did you get the number? After or inside which plugin?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.