Bounding box confidence filtering at NvDsInferParseCustomEfficientNMS and after

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hello everybody!

I am running a deepstream pipeline for inference with yolov7 and i am using NvDsInferParseCustomEfficientNMS to parse the output of the NMS plug in to deepstream format. I want to ask

  1. Where the NvDsInferParseCustomEfficientNMS takes the confidence from to filter (because for a reason it is 0.2 while the .trt is builded with 0 confidence)?
  2. Also i hardcoded the confidence there to be less than 0.2 and the bounding boxes seems to pass from this parser function, but i am not able to get them at at the tiler_sink_pad_buffer_prob where i extract the metadata.
    Can someone explain me how is this possible and if there are additional steps in between?

Thank you very much

3

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)

• DeepStream Version

• JetPack Version (valid for Jetson only)

• TensorRT Version

• NVIDIA GPU Driver Version (valid for GPU only)

• Issue Type( questions, new requirements, bugs)

• 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)

4

• Hardware Platform (Jetson / GPU) Jetson Xavier
• DeepStream Version DS 6.0
• JetPack Version (valid for Jetson only) 4.6 (L4T 32.6.1)
• TensorRT Version 8.0.1
**• Issue Type This is a question to better understand the inference pipeline pre and post processing and when the custom efficient nms parser is called (before the post process?) and which threshold it uses?

5

please refer to this code. the configuration “pre-cluster-threshold” will become that perClassThreshold.

1 Like