How to access object detection metadata from GStreamer pipeline with `qtimlvdetection`?

I’m working with GStreamer and the qtimlvdetection plugin for object detection using YOLO. I want to access the detected objects’ metadata, such as the label, confidence, and bounding box coordinates, and process it.

Currently, I’m using appsink to retrieve the metadata, but didn’t work properly

do you have other way to get metadata or Documents worth referencing

I’m trying it now by referring to the official GStreamer documentation and if I succeed, I’ll let you know.
I think I found a way.

Well, we look forward to hearing your positive news. :smiling_face_with_hearts:

Hi, I am working on the same thing, how did you manage to access this data?

I successfully manage to access the metadata.
You can refer to gst Buffer documentation.
and appsink.

If you need help contact me

You can refer to this site
https://mysupport.qualcomm.com/supportforums/s/question/0D7dK000002wbanSAA/detection-and-tracking-using-rb3gen2-device?

And I’m writing a Rubik’s Board development journal in Korean.

Thanks, I’ll give it a try and let you know how it went

I thought to edit the pipeline and use text/x-raw and then pass that to appsink, but I am not able to create such a pipeline. How did you do it?