DeepStream couldn't detect the targets properly

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1 
Jetpack 4.6.0
DeepStream 6.0.1
YOLOV5 6.1
torch 1.8
CUDA 10.2

My yolov5 model ( in pytorch ) was able to detect the target correctly, but when I using DeepStream detection, it couldn’t detect the targets properly, and there were so many false detection targets that almost occupied the entire screen.
I try to fix this problem by editing nms threshold , but it don’t work.

5

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

please refer to this deepstream yolov5 sample GitHub - beyondli/Yolo_on_Jetson