How find stopped people confidence instead of moving people on vehicle

[user89410]

Embeded device:jetson Nano 4gb
jetpack version:4.5.1
cuda version:10.2.89
SDK:deepstream-5.1
python code:deepstream python apps video analytics
ubuntu 18.04 LTS
TensorRT:7.1.3.1

In deepstream python apps deepstream-nvdsanalytics
problem statement : how can we get confidence of people…

confidence - Holds a confidence value for the object, set by the inference module.
confidence_of_people=obj_meta.confidence
The confidence value was negative. On doing some googlesearch, it was suggested
to set the cluster-mode=2 in dsnvanalytics_pgie_config.txt
On doing this, the people confidence was positive, but it affected the other detections.
The vehicle detection was incorrect.
How do we find the people/vehicle detection confidence.

regards
Gaurang Patel

[AastaLLL]

Hi,

First, please noted that we have some newer JetPack and Deepstream releases.
It’s recommended to upgrade your device to the latest JetPack 4.6.1 with Deepstream 6.0.1.

Below is an example to extract sgie confidence value based on deepstream-test2 example.
Please check if it can meet your requirement:

Python sample of chain a detection and classification models

Hi, 1. Usually, we use a tracker to generate the intermediate bounding box so you don’t need to apply the inference to every frame. Since Nano 2GiB is relatively resource-limited, enabling a tracker should be a good choice. 2. In the nvinfer configure, there is a filter-out-class-ids to filter out the unwanted class. https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvinfer.html#gst-nvinfer-file-configuration-specifications 3. Please try the following changes: diff…

Thanks.