Wrong result in bounding-boxes with custom sgie model

[minghua827]

**• Hardware Platform (Jetson / GPU)： GPU
**• DeepStream Version： 6.2
**• TensorRT Version： 8.5.1.7
**• NVIDIA GPU Driver Version (valid for GPU only)： 535
**• Issue Type( questions, new requirements, bugs)： questions
**• 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)

This is my pipeline: PGIE → SGIE → nvvidconv → nvosd → sink.

Both PGIE and SGIE are used as detectors, and I am not using the tracker plugin in between.

PGIE can detect two classes: CAN and HAND, but the results from SGIE are abnormal.

Could it be an issue with the NvDsInferParseCustomSGIEYoloV4 function (parse-bbox-func-name) ? How can I resolve this? Thanks.

Below are my relevant configuration files:

imagedata_pgie_config.txt (3.1 KB)

imagedata_sgie1_config.txt (3.9 KB)

NvDsInferParseCustomYoloV4.py.txt (5.9 KB)

[fanzh]

what is the sgie used to detect? is the sgie dependent on pgie? please refer to this sample deepstream_lpr_app, the pgie detects car, the sgie detects car license plate.

[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