Display confidence in the bounding box detected object

[soundarrajan]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) : Jetson Nano
• DeepStream Version : 6.0
• JetPack Version (valid for Jetson only) : 4.6
• TensorRT Version : 8.2
• NVIDIA GPU Driver Version (valid for GPU only) :
• Issue Type( questions, new requirements, bugs) : question
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Hi,

1. I want to display the confidence value of the detected object. Current label only showing with bbox. Expected: Label + confidence
   Is there any implemented example for this NvDsInfer API to refer?
   NvDsInfer — Deepstream Deepstream Version: 6.1.1 documentation

2. How can i print the FPS and inference timing in the output rendered video? Currently frame number only printing.

3. Is there any way to print or save raw metadata from gst_buffer? I tried,
   frame_meta = pyds.NvDsFrameMeta.cast(l_frame.data)
   print(‘\n-----printing frame meta-----’)
   print(frame_meta)
   It is printing only pointers actually…

[yuweiw]

Hi @soundarrajan , you can write anything or draw anything you want by the NvDsDisplayMeta struct.
You can refer deepstream_test1_app.c

static GstPadProbeReturn
osd_sink_pad_buffer_probe (GstPad * pad, GstPadProbeInfo * info,
    gpointer u_data)

First, get the data you wanted, and set it to the “display_meta”.
Thanks

[soundarrajan]

Hi @yuweiw,

I think you are not getting what i’m trying to say…
1) Expected output:

I can see Python API to get detectionConfidence, NvDsInferObjectDetectionInfo — Deepstream Deepstream Version: 6.1.1 documentation but i coudn’t see any sample implementation to refer.

2) FPS should be something like below image (FPS is printing in red color)

3) Metadata save
We are saving all data in GST_BUFFER and then probe in callback right?
I want to save the raw metadata whichever in the GST_BUFFER.

[soundarrajan]

Any updates on my queries?

Please let me know if any inputs are details required from my end…

[yuweiw]

Hi, @soundarrajan Do you run our demo from the link below or your own code?
https://github.com/NVIDIA-AI-IOT/deepstream_python_apps

The first one, Could you check from the python demo from the below link(cluster_and_fill_detection_output_nms):
https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/blob/master/apps/deepstream-ssd-parser/nms.py

The second one:
We have no the FPS and inference timing data statics now. In theory, you can draw any data that you get in the picture.
The third one:
You can refer https://forums.developer.nvidia.com/t/access-frame-pointer-in-deepstream-app/79838#5375214.

[yingliu]

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